Amygdala-lesion obesity: what is the role of the various amygdaloid nuclei?
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Am J Physiol Regulatory Integrative Comp Physiol
279: R1348–R1356, 2000.
Amygdala-lesion obesity: what is the role of the
various amygdaloid nuclei?
BETHANY L. ROLLINS AND BRUCE M. KING
Department of Psychology, University of New Orleans, New Orleans, Louisiana 70148
Received 2 December 1999; accepted in final form 24 May 2000
Rollins, Bethany L., and Bruce M. King. Amygdala- between the lateral and basal nuclei (20, 24, 50). Fon-
lesion obesity: what is the role of the various amygdaloid berg (19–22) found that lesions of the lateral nuclei,
nuclei? Am J Physiol Regulatory Integrative Comp Physiol particularly those in the ventral posterior portions,
279: R1348–R1356, 2000.—Anatomic descriptions of amyg- produced hyperphagia and obesity in dogs, whereas
daloid lesions resulting in hyperphagia and obesity in rats,
dorsomedial lesions (i.e., central and medial nuclei)
cats, and dogs have been inconsistent and often contradic-
tory, frequently resulting in failures to replicate. The present resulted in aphagia and weight loss. Wood (72) and
study attempted to reconcile these differences by examining Koikegami et al. (41), on the other hand, reported
common areas of overlap among differently placed lesions in marked hyperphagia in cats given lesions of the medial
female rats. Small bilateral lesions of the most posterodorsal and central nuclei. Anand and Brobeck (4) observed no
aspects of the amygdala resulted in substantial weight gains change in food intake in cats with lesions of either the
(mean ⫽ 45.4 g/10 days). The smallest lesions caused damage lateral amygdala or the central and medial amygdala.
limited to the posterodorsal medial amygdaloid nucleus and Numerous studies have examined the effects of
the bed nucleus of the stria terminalis and were directly in amygdaloid lesions on food intake and body weight in
the area where axons are collecting to form the stria termi- rats, resulting in a similar collection of inconsistent
nalis. Larger lesions that extensively damaged the central and contradictory results. Initial studies that em-
and/or anterodorsal medial amygdaloid nuclei sometimes
resulted in excess weight gains, as did very large lesions of
ployed large lesions that destroyed most of the amyg-
the basolateral nuclei, but substantial weight gains occurred dala reported hypophagia and weight loss (4, 12, 41,
only when the lesions extended (unilaterally or bilaterally) 56, 61, 64, 65). Subsequent studies of rats with smaller
into the posterodorsal amygdala. Examination of previously lesions aimed at the basolateral and/or lateral nuclei
published brain sections indicated that the hyperphagia and reported weight gain (8, 23, 43), no change in body
obesity that have been observed after widely differing lesion weight (18, 28, 40, 45, 58, 60), or even weight loss (14).
placements in cats and dogs were also the result of damage to Studies employing lesions of the corticomedial nuclei
a common area of overlap (i.e., the bed nucleus and/or stria were equally mixed, reporting weight gains (26),
terminalis). In rats, the critical area producing weight gain weight losses (12, 58), or no changes in body weight (59,
has extensive reciprocal relations with the medial hypothal- 60, 63). Central nucleus lesions have similarly been
amus. reported to result in either weight/fat gain (7, 42),
bed nucleus of the stria terminalis; stria terminalis aphagia and/or weight loss (8, 11, 23), or no change in
food intake and body weight (16, 29, 57, 59, 60). In
recent years, King and colleagues (30–33, 35–40) have
THE TEMPORAL LOBES have been suspected to play a role reported hyperphagia and moderate obesity in female
in feeding behavior ever since Brown and Schafer (9) rats given bilateral lesions of the most posterodorsal
reported in 1888 that temporal lobectomies in two aspects of the amygdala, with weight gains of 50–80 g
monkeys resulted in “insatiable” appetites. Later stud- typical in the first 15 to 20 days after lesions. The
ies confirmed hyperphagia as well as obesity in mon- critical nuclei were determined to be the posterodorsal
keys and humans with bilateral temporal lobe damage medial nucleus and the intra-amygdaloid bed nucleus
or temporal lobectomies (e.g., 10, 46, 55, 67), and sim- of the stria terminalis (G. F. Alheid, B. M. King, J. T.
ilar results were observed in monkeys given bilateral Cook, K. N. Rossiter, B. L. Rollins, S. J. Shammah-
resections of the amygdaloid complex and adjacent Lagnado, unpublished observations).
temporal cortex (62). This confusing array of results has deterred some
Subsequent lesion studies in dogs and cats estab- researchers from further exploring the role of the
lished that the critical site was indeed the amygdala amygdala in food intake and regulation of body weight
(20, 24, 41, 49, 50, 72). However, there was disagree- (personal communications). To better understand the
ment regarding the specific nuclei involved. Three effects of various amygdaloid lesions on body weight,
studies determined that the damage caused by obesity- the present study reexamined the effects of electrolytic
inducing lesions lie in the lateral nuclei or the junction
The costs of publication of this article were defrayed in part by the
Address for reprint requests and other correspondence: B. M. payment of page charges. The article must therefore be hereby
King, Dept. of Psychology, Univ. of New Orleans, New Orleans, LA marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
70148 (E-mail: bmking@uno.edu). solely to indicate this fact.
R1348 0363-6119/00 $5.00 Copyright © 2000 the American Physiological Society http://www.ajpregu.orgAMYGDALOID LESIONS AND BODY WEIGHT R1349
lesions of the amygdala in female rats. Common areas Formalin solution. The brains were stored in 10% Formalin
of overlap were examined to help identify the area of and later frozen and sliced into 40-m coronal sections. The
damage critical for weight gain. sections were stained with cresyl violet, and initial histolog-
ical analysis was performed in a blinded fashion (i.e., without
METHODS knowing changes in body weight) by light microscopic exam-
ination. The extent of the lesions was determined with use of
Animals. A total of 128 adult (110–130 days old) female the stereotaxic atlas by Paxinos and Watson (54).
Long-Evans hooded rats were used (Harlan Sprague-Dawley,
Indianapolis, IN). Females were used because the large ma-
RESULTS
jority of studies of the effects of ventromedial hypothalamic
lesions on food intake and body weight used female rats. All Sham lesions. Of the 29 rats with sham lesions
animals were individually housed in standard rat cages (9.5 observed in this study, 21 weighed less than they had
in. long ⫻ 7 in. wide ⫻ 7 in. high; no activity wheel) in a
temperature-controlled colony (21–24°C) with a 12:12-h
preoperatively on day 3 (mean change ⫾ SE ⫽ ⫺5.8 ⫾
light-dark cycle throughout the experiment. 1.1 g), and only 16 weighed more than they had preop-
Lesions. Bilateral electrolytic lesions were produced under eratively by the tenth day after surgery. The mean
pentobarbital sodium anesthesia (50 mg/kg) by passing an- 10-day weight change was ⫹1.7 ⫾ 1.8 g, and the great-
odal current between the uninsulated tip of an insulated est weight gain observed was 18 g (see Table 1).
stainless steel electrode (Plastics One, Roanoke, VA) and a Posterodorsal lesions. “Posterodorsal” refers to an
rectal cathode. Posterodorsal lesions and central nucleus area, not a specific nucleus (analogous to ventromedial
lesions were performed with a 1.5-mA current produced for hypothalamus vs. ventromedial hypothalamic nuclei).
20 s with 0.2 (posterodorsal) or 0.4 mm (central) of the Four serial sections of six lesions are provided in Fig. 1.
electrode tip uninsulated. Basolateral amygdaloid lesions In terms of weight gain, the most effective lesions were
were produced with a 2.0-mA current for 30 s with 0.9 mm of
the electrode tip uninsulated. Electrodes were positioned
centered at their maximal point of damage immedi-
relative to bregma with the use of a Kopf small animal ately ventral and adjacent to the dorsal tip of the optic
stereotaxic instrument. With the upper incisor bar positioned tract in the posterior aspects of the amygdala. With
horizontally with the interaural line, the electrodes for pos- small lesions (of which all 6 in Fig. 1 are examples), the
terodorsal lesions were positioned 1.7 mm posterior to posterior extent of the damage fuses into (and eventu-
bregma (AP), 4.5 mm lateral to the midsaggital suture (ML), ally becomes indistinguishable from) the lateral ven-
and 8.4 mm below the surface of the skull (DV). For basolat- tricle. In the case of larger lesions or lesions that begin
eral lesions, the coordinates were (in mm) AP ⫺0.8, ML 4.7, more posterior than those in Fig. 1, the damage nor-
and DV 9.4. For lesions involving a greater portion of the mally extends into the amygdalohippocampal area and
central nucleus, the coordinates were (in mm) AP ⫺0.4, ML ventral hippocampus. However, these lesions are less
4.2, and DV 8.4. Control animals had holes drilled in the
skull at the same coordinates, and electrodes were lowered to
effective in producing weight gain (see Ref. 30). Weight
a depth 1.0 mm above the target site. gain was also attenuated whenever the damage ex-
Procedure and histology. The study compared four groups tended into the ventral aspects of the caudal globus
of animals: sham lesions (n ⫽ 29), posterodorsal amygdaloid pallidus.
lesions (n ⫽ 47), basolateral amygdaloid lesions (n ⫽ 37), and An examination of Fig. 1 reveals that lesions in the
lesions aimed at the central nucleus (n ⫽ 15). Rats with most posterodorsal aspects of the amygdala do not
sham, posterodorsal, or basolateral lesions were used in a have to be large to produce weight gain. For example,
subsequent behavioral study (to be reported separately), rat 1F sustained the smallest lesions yet displayed the
which required that 21 of the animals with posterodorsal largest weight gain (63 g/10 days). The lesion illus-
lesions be food restricted starting on day 4. Except for being trated in Fig. 1A is also remarkable for its small size
weighed daily, all the other animals (including 26 rats with
posterodorsal lesions) were undisturbed for 10 days. All an-
and the substantial weight gain it produced. Histolog-
imals were allowed to feed ad libitum (Harlan Teklad pellet- ical analysis of 32 rats sustaining small lesions that
form rat diet LM-485) during the 3- or 10-day observation produced substantial weight gains revealed that the
period reported here. areas of damage they shared in common were in the
At completion of the study, the rats with lesions were posterodorsal medial amygdaloid nucleus and the bed
killed and perfused with physiological saline and a 10% nucleus of the stria terminalis. There were varying
Table 1. Initial body weights, weight changes, and daily food intake of female rats with amygdaloid lesions
Food Intake Weight Change
Group n Initial Weight Days 1–3 Days 4–10 Day 3 Day 10
Sham lesions 29 279.2 ⫾ 3.4 18.9 ⫾ 0.6 22.3 ⫾ 0.6 ⫺5.8 ⫾ 1.1 1.7 ⫾ 1.8
PDA 11 271.0 ⫾ 3.8 31.0 ⫾ 0.3 34.6 ⫾ 1.2 22.8 ⫾ 1.9 45.4 ⫾ 2.4
Basolateral nuclei
Basolateral 12 275.4 ⫾ 3.4 22.3 ⫾ 2.0 25.2 ⫾ 1.3 0.8 ⫾ 1.6 9.8 ⫾ 2.6
Basolateral ⫹ partial PDA 5 287.2 ⫾ 2.4 26.9 ⫾ 2.3 31.9 ⫾ 2.5 0.0 ⫾ 6.2 21.0 ⫾ 1.4
Basolateral ⫹ PDA 3 280.0 ⫾ 9.4 30.2 ⫾ 5.5 33.5 ⫾ 4.8 16.7 ⫾ 8.6 32.0 ⫾ 7.1
Central nuclei ⫹ PDA 7 288.6 ⫾ 7.1 31.5 ⫾ 4.0 30.6 ⫾ 2.0 16.8 ⫾ 3.1 22.7 ⫾ 3.6
Values are mean grams ⫾ SE. Posterodorsal area (PDA) data are only for those rats with small bilateral lesions that were fed ad libitum
for 10 days.R1350 AMYGDALOID LESIONS AND BODY WEIGHT Fig. 1. Four serial sections through the brains of 6 rats with small lesions in the posterodorsal area of the amygdala. The lesions were centered in the posterodorsal medial amygdaloid nucleus and bed nucleus of the stria terminalis and then fused with the lateral ventricle posteriorally. Note also that the lesions directly damage the site at which axons come together to form the stria terminalis. degrees of damage to the medial and dorsal portions of stria terminalis, and eventual fusion with the lateral the posterior basomedial nucleus and the caudal por- ventricle. However, all three lesions invaded the ven- tions of the central nucleus. The lesions were also in tral aspects of the globus pallidus to varying degrees. the immediate area where axons come together to form This was especially true of the lesions in rat B. Despite the stria terminalis. Examination of the smallest le- this, the weight gain for the seven animals with bilat- sions (e.g., Fig. 1, A and F) revealed that there was eral damage to the central nuclei (and varying degrees greater damage to the bed nucleus than to the pos- of damage to the posterodorsal amygdala) still ex- terodorsal medial nucleus. For 11 rats with very small ceeded those observed after sham lesions (mean ⫽ bilateral lesions that were fed ad libitum for 10 days, 22.7 ⫾ 3.6 g/10 days). the mean weight gain was 22.8 ⫾ 1.9 g after 3 days and Rat 2D had the greatest bilateral damage to the 45.4 ⫾ 2.4 g after 10 days. Daily food intakes reflected medial nucleus of any rat observed. The damage in- weight gains (see Table 1). cluded the anterodorsal division of the medial nucleus Central and medial lesions. Seven of fifteen animals in anterior sections, the posterodorsal division and bed had extensive bilateral damage to the central nucleus. nucleus of the stria terminalis in posterior sections, Four serial sections through the brains of three of these and complete fusion with the lateral ventricle. There rats are displayed in Fig. 2 (A, B, C). Note that the was no damage to the globus pallidus. However, weight lesions are larger than those observed in rats with gain (34 g/10 days) was no greater in this rat than in posterodorsal lesions (Fig. 1). The lesions for rats A, B, rats with much smaller lesions limited to the pos- and C began more anterior than the lesions displayed terodorsal medial amygdaloid nucleus and bed nucleus in Fig. 1 and included much more extensive damage to (Fig. 1). the central nucleus. But because the lesions were Basolateral lesions. Because of the amount of elec- larger, the damage continued posterior to include trode tip exposed and duration of current, lesions much of the area destroyed by posterodorsal lesions, aimed at the basolateral nuclei were large. However, i.e., posterodorsal medial nucleus, bed nucleus of the there was considerable variation in the extent of the
AMYGDALOID LESIONS AND BODY WEIGHT R1351 Fig. 2. Four serial sections through the brains of 3 rats with lesions of the central amygdaloid nucleus that included extensive damage to the posterodorsal amygdaloid area (A, B, C) and a 4th rat with extensive damage to the anterodorsal medial amygdaloid nucleus (D). damage. Of 37 animals with lesions, 12 had lesions sphere) to damage large portions of the posterodorsal that, either unilaterally or bilaterally, extended well medial nucleus and the bed nucleus of the stria termi- lateral to (and greatly spared) the basolateral nuclei. nalis (e.g., Fig. 3C). The mean 10-day weight gain for Another three had lesions that were too posterior, and these five rats was 21 g. in two others the damage spared over half of the nuclei For the remaining 12 rats, mean 3- and 10-day ventrally. The data for these 17 animals were elimi- weight gains were 0.8 ⫾ 1.6 and 9.8 ⫾ 2.6 g, respec- nated, leaving 20 animals with extensive bilateral tively. The 10-day weight gains (range ⫽ ⫺4 to ⫹23 g) damage to the basolateral nuclei. However, among were significantly greater than those observed in rats these 20 animals, there was still considerable variation with sham lesions (t ⫽ 2.52, df ⫽ 39, P ⬍ 0.01, d ⫽ in the dorsal-ventral extent of the damage, with vary- 0.86). However, even among this group, the best weight ing amounts of damage to the lateral, basolateral, and gains were observed in those animals in which lesions basomedial amygdaloid nuclei. Brain sections of eight were large enough to extend far enough posterior to representative rats with basolateral lesions are pre- infringe on the most posterior aspects of the bed nu- sented in Fig. 3. cleus just ventral to the lateral ventricle. The 10-day weight changes in the 20 rats with bilat- eral basolateral lesions ranged from ⫺4 to ⫹44 g. In DISCUSSION the initial examination of the brain sections, it was immediately apparent that weight gain was directly The weight changes that were observed in female related to the dorsal-most extent of the damage in any rats given sham lesions were similar to those that had one hemisphere. Sections were enlarged with a micro- previously been reported. Adult female rats almost projector, and the distance measured between 1) the always lose weight in the first 3 to 5 days after sham most dorsal point of the lesion and 2) the dorsal tip of lesions, and many do not exceed their preoperative the optic tract within the same coronal plane in which weights by day 10 (30, 33, 35–37, 40). Against this the most effective posterodorsal lesions were observed. backdrop, the initial weight gain of female rats with There was a significant negative correlation between posterodorsal amygdaloid lesions is impressive. Al- this distance and 10-day weight gain (r ⫽ ⫺0.81, df ⫽ though female Long-Evans rats generally take 6 to 19, P ⬍ 0.001, r2 ⫽ 0.66). For three of the rats with 10 h to become fully ambulatory after pentobarbital basolateral lesions, damage to the posterodorsal amyg- sodium anesthesia, rats with posterodorsal lesions of- dala was nearly identical (unilaterally or bilaterally) to ten display a weight gain in the first 24 h after surgery. that observed in rats with intentional posterodorsal In the present study, the initial mean 3-day gain of lesions (e.g., Fig. 3, A and B). Their mean 10-day 22.8 g computes to a net difference of ⫹28.6 g compared weight gain was 32 g. Five additional rats had damage with the mean loss of ⫺5.8 g displayed by rats with that extended far enough dorsally (in at least 1 hemi- sham lesions. The 3- and 10-day weight gains of the
R1352 AMYGDALOID LESIONS AND BODY WEIGHT Fig. 3. Coronal sections of 8 rats with large lesions of the basolateral nuclei. The rats are displayed in order from top to bottom in terms of extent of damage to the posterodorsal amygdala in either hemisphere. rats with posterodorsal lesions were comparable to the et al. (unpublished observations) used multiple regres- gains that have often been reported for female rats sion analysis to determine that damage to the intra- with lesions of the paraventricular hypothalamic nu- amygdaloid bed nucleus of the stria terminalis and the cleus (5, 68–70). The weight gains were less than those posterodorsal medial amygdaloid nucleus accounts for generally observed after lesions of the ventromedial ⬃35% of the variance in weight gain after (larger) hypothalamus (e.g., 34). Weight gain in rats with pos- posterodorsal amygdaloid lesions. Examination of the terodorsal amygdaloid lesions generally plateaus 15 to smallest lesions in the present study revealed that 20 days after surgery, compared with 3 to 6 wk for rats there was little damage to other amygdaloid nuclei, with well-placed lesions of the ventromedial hypothal- and examination of the lesion of rat D in Fig. 2 suggests amus (34). that more extensive damage to the medial nucleus does Analysis of the brains of rats with the smallest not further enhance weight gain. Others have also posterodorsal lesions indicated that the critical area for observed no weight gain in rats with lesions of the weight gain included the posterodorsal medial amyg- (posteroventral and anterodorsal) medial amygdaloid daloid nucleus and the bed nucleus of the stria termi- nuclei and/or cortical nuclei (28, 59, 60, 63). Alheid et nalis. Damage to these structures can clearly be seen al. (unpublished observations) found that damage to in the serial sections of Fig. 1. In a recent study, Alheid the caudal globus pallidus was negatively loaded in
AMYGDALOID LESIONS AND BODY WEIGHT R1353
predicting the weight gain (i.e., damage reduced rats with lesions that included damage to the ventral
weight gain). Together, damage to the posterodorsal half of the central nuclei.
medial amygdaloid nucleus, the intra-amygdaloid bed Ganaraj and Jeganathan (23) also reported that ba-
nucleus, and the globus pallidus accounted for 97% of solateral lesions resulted in small weight gains
the variability in weight gain. (mean ⫽ 23 g/3 wk) in young male rats, but their data
Although damage produced by the smallest lesions are equally notable for a lack of male-typical weight
was limited to the posterodorsal medial amygdaloid gain by rats with sham lesions (mean ⫽ 2 g/3 wk). In
nuclei and bed nucleus of the stria terminalis, one our experience, young male rats with sham lesions
must also keep in mind that axons are collecting begin to display sex-typical weight gains within days
throughout these areas to form the stria terminalis. after surgery (37). Most studies that have examined
Three studies reported that sectioning of the stria did the effects of bilateral electrolytic lesions of the baso-
not result in excessive weight gain in male rats (6, 8, lateral nuclei in male rats did not find excessive weight
51), but these results may have been due to a sex gains (e.g., 14, 18, 45, 58, 60), although Box and Mo-
difference in weight gain after amygdaloid lesions (37) genson (8) reported hyperphagia and obesity in four
and damage to the globus pallidus (6, 8), internal male rats with lesions of the ventral basolateral nuclei
capsule (8), or cortex and dorsal striatum (51). Ehrlich in the most posterior aspects of the amygdala. Lenard
(17) reported excessive weight gains in male rats given et al. (43) observed a small increase in weight gain in
electrolytic lesions of the fornix (26% greater than male rats given 6-OHDA lesions of the lateral nuclei
controls in the first 3 days), but an examination of the (⬃130% of preoperative weight compared with 124–
lesions she made indicates that the damage was large 125% for controls after 18 days).
enough to have almost certainly included the stria At first glance, the results obtained with basolateral
terminalis. Coronal knife cuts anterior to the ventro- lesions in the present study appeared to support those
medial hypothalamic nuclei, a projection site for a of Ganaraj and Jeganathan (23). However, the lesions
major branch of the stria terminalis, have been found in the present study were quite large, and there was a
to result in hyperphagia and/or excessive weight gain high correlation between weight gain and the dorsal
in female rats and some male rats (25, 53, 66). Food extent of the lesions. The most substantial weight
intake was at least 150% of normal, and weight gains gains were observed in rats with lesions that clearly
invaded the posterodorsal medial amygdaloid nucleus
were at least double the normal rate. In addition,
and the bed nucleus of the stria terminalis dorsally in
electrical stimulation of the medial portion of the
at least one hemisphere. Previous work has estab-
amygdala suppresses food intake, and transection of
lished that unilateral posterodorsal amygdaloid lesions
the stria terminalis prevents the suppression (71). Ex-
result in weight gains equal to those produced by
amination of the effects of injection of a cellular neu-
bilateral lesions during the first 2–3 days after surgery
rotoxin (e.g., ibotenic acid) into the posterodorsal (32). [Unilateral ventromedial hypothalamic lesions
amygdala should be valuable in clarifying whether the also result in hyperphagia and obesity (47).] Neverthe-
effects of electrolytic lesions are due to damage to local less, when the data for these rats (e.g., Fig. 3, A, B, and
nuclei or to fiber pathways. C) were eliminated, the remaining 12 animals still
The finding that incidental damage to the globus displayed a small but statistically significant increase
pallidus is negatively correlated with weight gain (Al- in weight gain compared with controls (9.8 vs. 1.7 g,
heid et al., unpublished observations) is particularly respectively, in 10 days). However, even among these
pertinent when considering the role of the central nu- rats, weight gain was clearly related to the dorsal
cleus. Box and Mogenson (8) and Ganaraj and Jega- extent of the lesions. Compare, for example, the lesions
nathan (23), for example, reported that central nucleus of rats D and E (Fig. 3) with those of rats G and H. All
lesions “drastically” decreased food intake and body four have extensive damage to the basolateral nuclei,
weight. However, besides aphagia, their rats displayed but the lesions in rats D and E (which resulted in
sensory-motor deficits, including excessive gnawing moderate weight gain) have a greater dorsal extension.
and spillage of their food pellets. Others have reported Rats G and H displayed no excessive weight gains. This
similar results for rats, cats, and dogs (e.g., 4, 11, 19, suggests that the small weight gains observed in these
21). Aphagia is a well-established consequence of dam- 12 animals may have been due to the damage infring-
age to the caudal globus pallidus (16, 44, 48, 52, 60). ing on the critical areas within the posterodorsal amyg-
Previous studies had already established that lesions dala (i.e., posterodorsal medial nucleus/bed nucleus or
limited to the central nucleus, without incidental dam- ventral portion of the area where axons are coming
age to the globus pallidus, do not result in hypophagia, together to form the stria terminalis). Two previous
weight loss, or sensory-motor deficits (7, 16, 29, 57, 60, studies reported hyperphagia in rats with lesions of the
see also Ref. 42). Obesity-inducing posterodorsal amyg- ventral amygdala, but the effects were either very
daloid lesions often invade the caudal portions of the small (26) or transient [and observed only on a high-fat
central nucleus, and the present results demonstrate diet (8)]. The lesions in another study that resulted in
that excessive weight gains are possible even in rats mild hyperphagia were posterior and extended dor-
with much more extensive damage to the central nu- sally into the same region described here (15).
cleus (Fig. 2, A, B, and C). Box and Mogenson (8) also The excessive weight gains observed in the present
observed mild hyperphagia and weight gain in male study after a variety of differently placed amygdaloidR1354 AMYGDALOID LESIONS AND BODY WEIGHT
lesions (posterodorsal area, medial, basolateral, and tral nuclei. The best evidence favors the basolateral or
even central nucleus) may be important in reconciling lateral nuclei. However, it should be noted that in
the discrepant results reported for other species. In those few studies that found hyperphagia and/or exces-
dogs and cats, dorsomedial lesions (medial and central sive weight gain in rats with basolateral or lateral
nuclei) have resulted in both aphagia/weight loss (4, lesions, the effects were usually very small and the
19, 21) and hyperphagia/obesity (41, 72), whereas le- lesions were either very dorsal or very posterior (8, 15,
sions of the lateral or basolateral group of nuclei have 23, 26, 43, present study).
resulted in no change (4) or hyperphagia and obesity In conclusion, the present study identifies the critical
(20, 21, 24, 49, 50). Inspection of published photo- site for lesion-induced obesity in rats to be the pos-
graphs of lesions suggests that the aphagia following terodorsal medial amygdaloid nucleus and the bed
lesions of the medial and central nuclei were probably nucleus of the stria terminalis, as indicated by both
due to additional damage to adjacent nonamygdaloid very small lesions and as the point of overlap in vari-
structures involved in sensory-motor functions [e.g., 19 ously placed large lesions (when there is no incidental
(Fig. 4, p. 742); 21 (Fig. 6, p. 458)]. damage to the globus pallidus). This is the same area
Regarding hyperphagia and obesity, two possible in which axons come together to form the stria termi-
explanations of the role of the amygdala arise: 1) much nalis. Examination of obesity-inducing lesions in cats
of the amygdala is inhibitory for feeding behavior and and dogs also suggests that the critical area is the bed
body weight or 2) the seemingly diverse lesions over- nucleus and/or stria terminalis. With the determina-
lapped in some critical area. The first possibility seems tion of the critical amygdaloid site for lesion-induced
unlikely in view of the vastly different embryonic de- obesity, future research can now be directed at under-
velopment of the medial/central nuclei vs. the basolat- standing the etiology of this experimentally induced
eral complex of nuclei (27). The results of the present obesity.
study with rats support the second possibility. Sub-
stantial weight gains were observed only when lesions Perspectives
infringed on the most posterodorsal aspects of the
amygdala, regardless of whether the lesions were cen- Lesions of the posterodorsal amygdala in many ways
tered in the basolateral group of nuclei, the medial mimic, although to a smaller extent, lesions of the
nucleus, or the central nucleus. Published photographs ventromedial hypothalamus. This includes dynamic
of brain sections for other species are limited, and one and static phases of hyperphagia and weight gain (33),
must also keep in mind the medial rotation of the hyperinsulinemia (31), and a greater magnitude of
amygdala as species ascend phylogenetically (27). weight gain in females (37). Lesions of the posterodor-
However, an inspection of the basolateral lesions in sal amygdala in rats do, in fact, result in extensive
cats [24 (Figs. 17 and 18, p. 543)] and lateral lesions in anterograde degeneration in the stria terminalis and
dogs [22 (Fig. 5b, p. 73)] reveals extension dorsally into ventromedial hypothalamic nuclei, with little or no
a common area near the dorsal tip of the optic tract. degeneration in the paraventricular hypothalamic nu-
The lesions spared the medial nucleus, but the fibers of clei (Alheid et al., unpublished observations; 32). Ini-
the stria terminalis still collect in this area. In fact, in tial studies of putative neurotransmitter mechanisms
the figure legend of their photographs, Green et al. (24) suggest a possible role for serotonin (13). Alheid and
state that in addition to damage to the lateral nucleus, colleagues (1, 2) include the posterodorsal amygdala as
the ventral part of the stria terminalis is injured. part of the medial corridor of the “extended amygdala.”
Lesions that did not result in weight gain were clearly However, posterodorsal amygdaloid lesions do differ
more ventral to this [24 (Figs. 19 and 20, p. 543)], notably from ventromedial hypothalamic lesions in
similar to G and H in Fig. 3 of the present study. Wood some respects. This includes a strong preference for
(72) did not publish photographs, but if the lesions carbohydrates (39), but otherwise a lack of finickiness
involved the central nucleus as described, they, too, (38). Thus there is no exact duplication of feeding systems
would have damaged the bed nucleus and stria termi- within the amygdala and medial hypothalamus. It is
nalis. Morgane and Kosman’s [50 (Fig. 1, p. 159)] suggested that the amygdala modulates the hypothala-
lesions were large and destroyed nearly all of the mus with regards to regulation of food intake. More
amygdala. Koikegami et al. [41 (Fig. 1, p. 215)] ob- succinctly, the ventromedial hypothalamus, with its nu-
served obesity in only one cat, but this medially placed merous glucose receptors, may monitor “hunger,”
lesion also infringed on the proposed critical area at whereas the amygdala, with its major afferent input from
the dorsal end of the optic tract. the olfactory bulb, may direct goal-oriented “feeding be-
We cannot rule out the possibility that lesions that havior.”
spare the posterodorsal amygdala (i.e., posterodorsal
medial nucleus, bed nucleus, and stria terminalis) re- REFERENCES
sult in small weight gains. In fact, should the critical 1. Alheid GF, de Olmos J, and Beltramino CA. Amygdala and
structure prove to be the stria terminalis, then one extended amygdala. In: The Rat Nervous System, edited by
would expect to observe some weight gain after lesions Paxinos G. San Diego, CA: Academic, 1995, p. 495–578.
2. Alheid GF and Heimer L. Theories of basal forebrain organi-
of nuclei from which axons of the stria terminalis zation and the “emotional motor system.” In: Progress in Brain
originate. There is little evidence to support the corti- Research, edited by Holstege G, Bandler R, and Saper CB.
cal (posteroventral and anterodorsal), medial, or cen- Amsterdam: Elsevier, 1995, p. 461–484.AMYGDALOID LESIONS AND BODY WEIGHT R1355
4. Anand BK and Brobeck JR. Food intake and spontaneous exploration and taste aversion. Physiol Behav 22: 789–793,
activity of rats with lesions in the amygdaloid nuclei. J Neuro- 1979.
physiol 15: 421–430, 1952. 30. King BM, Arceneaux ER, Cook JT, Benjamin AL, and
5. Aravich PE and Sclafani A. Paraventricular hypothalamic Alheid GF. Temporal lobe lesion-induced obesity in rats: an
lesions and medial hypothalamic knife cuts produce similar anatomical investigation of the posterior amygdala and hip-
hyperphagia syndromes. Behav Neurosci 97: 970–983, 1983. pocampal formation. Physiol Behav 59: 843–848, 1996.
6. Black RM and Weingarten HP. A comparison of taste reac- 31. King BM, Cook JT, and Dallman MF. Hyperinsulinemia in
tivity changes induced by ventromedial hypothalamic lesions rats with obesity-inducing amygdaloid lesions. Am J Physiol
and stria terminalis transections. Physiol Behav 44: 699–708, Regulatory Integrative Comp Physiol 271: R1156–R1159, 1996.
1988. 32. King BM, Cook JT, Rossiter KN, Sam H, Arceneaux E, and
7. Bovetto S and Richard D. Lesion of central nucleus of amyg- Alheid GF. Obesity after unilateral amygdala lesions in rats: a
dala promotes fat gain without preventing effect of exercise on quantitative analysis. Soc Neurosci Abstr 23: 251, 1997.
energy balance. Am J Physiol Regulatory Integrative Comp 33. King BM, Cook JT, Rossiter KN, Theobold LM, and Sam
Physiol 269: R781–R786, 1995. HM. Posterodorsal amygdaloid lesions in rats: long-term effects
8. Box BM and Mogenson GJ. Alterations in ingestive behaviors on body weight. Physiol Behav 60: 1569–1571, 1996.
after bilateral lesions of the amygdala in the rat. Physiol Behav 34. King BM and Gaston MG. Reappearance of dynamic hy-
15: 679–688, 1975. perphagia during the static phase in medial hypothalamic le-
9. Brown S and Schafer EA. An investigation into the functions
sioned rats. Physiol Behav 18: 945–950, 1977.
of the occipital and temporal lobes of the monkey’s brain. Philos
35. King BM, Kass JM, Cadieux NL, Sam H, Neville KL, and
Trans R Soc Lond B Biol Sci 179: 303–327, 1888.
Arceneaux ER. Hyperphagia and obesity in female rats with
10. Bucy PC and Kluver H. An anatomical investigation of the
temporal lobe lesions. Physiol Behav 54: 759–765, 1993.
temporal lobe in the monkey (Macacca mulatta). J Comp Neurol
103: 151–251, 1955. 36. King BM, Kass JM, Neville KL, Sam H, Tatford AC III, and
11. Cole SO. Dorsomedial amygdala damage: a time-after-surgery Zansler CA. Abnormal weight gain in rats with amygdaloid
assessment of feeding. Bull Psychonom Soc 9: 399–401, 1977. lesions. Physiol Behav 54: 467–470, 1993.
12. Collier BD and Gault FP. Aphagia and adipsia following 37. King BM, Rollins BL, Stines SG, Cassis SA, McGuire HB,
lesions of the amygdala. Psychonom Sci 17: 41–42, 1969. and Lagarde ML. Sex differences in body weight gains follow-
13. Coscina DV, Currie PJ, Bishop C, Parker GC, Rollins BL, ing amygdaloid lesions in rats. Am J Physiol Regulatory Integra-
and King BM. Posterodorsal amygdala lesions blunt feeding tive Comp Physiol 277: R975–R980, 1999.
induced by 8-OH-DPAT. Soc Neurosci Abstr 25: 2140, 1999. 38. King BM, Rossiter KN, Cook JT, and Sam HM. Amygdaloid
14. Crow TJ and Whitaker IM. A short-term effect of amygdaloid lesion-induced obesity in rats in absence of finickiness. Physiol
lesions on food intake in the rat. Exp Neurol 27: 520–526, 1970. Behav 62: 935–938, 1997.
15. Czech DA. Effects of amygdalar lesions on eating and drinking 39. King BM, Rossiter KN, Stines SG, Zaharan GM, Cook JT,
and saline preference in the rat. Physiol Behav 10: 821–823, Humphries MD, and York DA. Amygdaloid-lesion hyperpha-
1973. gia: impaired response to caloric challenges and altered macro-
16. Dacey DM and Grossman SP. Aphagia, adipsia, and sensory- nutrient selection. Am J Physiol Regulatory Integrative Comp
motor deficits produced by amygdala lesions: a function of extra- Physiol 275: R485–R493, 1998.
amygdaloid damage. Physiol Behav 19: 389–395, 1977. 40. King BM, Sam H, Arceneaux ER, and Kass JM. Effect on
17. Ehrlich A. Effects of tegmental lesions on motivated behavior in food intake and body weight of lesions in and adjacent to the
rats. J Comp Physiol Psychol 56: 390–396, 1963. posterodorsal amygdala in rats. Physiol Behav 55: 963–966,
18. Fitzgerald RE and Burton MJ. Effects of small basolateral 1994.
amygdala lesions on ingestion in the rat. Physiol Behav 27: 41. Koikegami H, Fuse S, Hiroki S, Kazami T, and Kageyama
431–437, 1981. Y. On the inhibitory effect upon the growth of infant animals or
19. Fonberg E. Effects of small dorsomedial amygdala lesions on on the obesity in adult cat induced by bilateral destruction of the
food intake and acquisition of instrumental alimentary reactions amygdaloid nuclear region. Folia Psychiatr Neurol Jpn 12: 207–
in dogs. Physiol Behav 4: 739–743, 1969. 223, 1958.
20. Fonberg E. Hyperphagia produced by lateral amygdala lesions 42. Lenard L and Hahn Z. Amygdalar noradrenergic and
in dogs. Acta Neurobiol Exp (Warsz) 31: 19–32, 1971. domaminergic mechanisms in the regulation of hunger and
21. Fonberg E. The normalizing effect of lateral amygdalar lesions thirst-motivated behavior. Brain Res 233: 115–132, 1982.
upon the dorsomedial amygdalar syndrome in dogs. Acta Neuro- 43. Lenard L, Hahn Z, and Karadi Z. Body weight changes after
biol Exp (Warsz) 33: 449–466, 1973. neurochemical manipulations of lateral amygdala: noradrener-
22. Fonberg E. The relation between alimentary and emotional gic and dopaminergic mechanisms. Brain Res 249: 95–101, 1982.
amygdalar regulation. In: Hunger: Basic Mechanisms and Clin- 44. Lenard L, Sarkisian J, and Szabo I. Sex-dependent survival
ical Implications, edited by Novin D, Wyrwicka W, and Bray G.
of rats after bilateral pallidal lesions. Physiol Behav 15: 389–
New York: Raven, 1976, p. 61–75.
397, 1975.
23. Ganaraj B and Jeganathan PS. Involvement of basolateral
45. Lorenzini CA, Baldi E, Bucherelli C, Giachetti A, and
nucleus & central nucleus of amygdala in the regulation of
Tassoni G. Effects of nucleus basolateralis amygdalae neuro-
ingestive behaviour in rat. Indian J Med Res 108: 98–103, 1998.
24. Green JD, Clemente CD, and de Groot J. Rhinencephalic toxic lesions on some spontaneous activities in the rat. Physiol
lesions and behavior in cats. J Comp Neurol 108: 505–545, 1957. Behav 50: 1215–1219, 1991.
25. Grossman SP. Changes in food and water intake associated 46. Marlowe WB, Mancall EL, and Thomas JJ. Complete Klu-
with an interruption of the anterior or posterior fiber connections ver-Bucy syndrome in man. Cortex 11: 53–59, 1975.
of the hypothalamus. J Comp Physiol Psychol 75: 23–31, 1971. 47. Mayer J and Barnett RJ. Obesity following unilateral hypo-
26. Grossman SP and Grossman L. Food and water intake fol- thalamic lesions in rats. Science 121: 599–600, 1955.
lowing lesions or electrical stimulation of the amygdala. Am J 48. Morgane PJ. Alteration in feeding and drinking behavior of
Physiol 205: 761–765, 1963. rats with lesions in the globi pallidi. Am J Physiol 201: 420–428,
27. Humphrey T. The development of the human amygdaloid com- 1961.
plex. In: The Neurobiology of the Amygdala, edited by Elefthe- 49. Morgane PJ and Kosman AJ. Alterations in feline behaviour
riou BE. New York: Plenum, 1972, p. 21–77. following bilateral amygdalectomy. Nature 180: 598–600, 1957.
28. Kemble ED and Godding PR. Effects of basolateral or corti- 50. Morgane PJ and Kosman AJ. A rhinencephalic feeding center
comedial amygdaloid lesions on grooming, consummatory, and in the cat. Am J Physiol 197: 158–162, 1959.
locomotor behaviours in rats. Behav Processes 6: 161–169, 1981. 51. Myhrer T. Effects of stria terminalis sections on locomotor,
29. Kemble ED, Studelska DR, and Schmidt MK. Effects of avoidance, and alternation behavior in rats. Physiol Psychol 3:
central amygdaloid nucleus lesions on ingestion, taste reactivity, 245–248, 1975.R1356 AMYGDALOID LESIONS AND BODY WEIGHT
52. Palfai T, Armstrong D, and Courtney CL. Effect of L-dopa or 63. Sclafani A, Belluzzi JD, and Grossman SP. Effects of lesions
bromocriptine on feeding and motor behavior of rats with lesions in the hypothalamus and amygdala on feeding behavior in the
in the globus pallidus. Physiol Behav 33: 283–289, 1984. rat. J Comp Physiol Psychol 72: 394–403, 1970.
53. Paxinos G and Bindra D. Hypothalamic knife cuts: effects on 64. Stoller WL. Effects of septal and amygdaloid lesions on discrim-
eating, drinking, irritability, aggression, and copulation in the ination, eating and drinking. Physiol Behav 8: 823–828, 1972.
male rat. J Comp Physiol Psychol 79: 219–229, 1972. 65. Stoller WL and Stoller RAV. Eating and drinking in rats with
54. Paxinos G and Watson C. The Rat Brain in Stereotaxic Coor- anterior or posterior amygdaloid lesions. Bull Psychonom Soc 11:
dinates (4th ed.). San Diego, CA: Academic, 1998. 43–45, 1978.
55. Pribram KH and Bagshaw M. Further analysis of the tempo- 66. Storlien LH and Albert DJ. The effect of VMH lesions, lateral
ral lobe syndrome utilizing frontotemporal ablations. J Comp cuts and anterior cuts on food intake, activity level, food moti-
Neurol 99: 347–375, 1953. vation, and reactivity to taste. Physiol Behav 9: 191–197, 1972.
56. Pubols LM. Changes in food-motivated behavior of rats as a 67. Terzian H and Ore GD. Syndrome of Kluver and Bucy repro-
function of septal and amygdaloid lesions. Exp Neurol 15: 240– duced in man by bilateral removal of the temporal lobes. Neu-
254, 1966. rology 5: 373–380, 1955.
57. Ritter S and Hutton B. Mercaptoacetate-induced feeding is 68. Tokunaga K, Fukushima M, Kemnitz JW, and Bray GA.
impaired by central nucleus of the amygdala lesions. Physiol Comparison of ventromedial and paraventricular lesions in rats
Behav 58: 1215–1220, 1995. that become obese. Am J Physiol Regulatory Integrative Comp
58. Rolls ET and Rolls BJ. Altered food preferences after lesions in Physiol 251: R1221–R1227, 1986.
the basolateral region of the amygdala in the rat. J Comp Physiol 69. Tokunaga K, Fukushima M, Lupien JR, Bray GA, Kemnitz
Psychol 83: 248–259, 1973. JW, and Schemmel R. Effects of food restriction and adrenal-
59. Rosen EF. Amygdaloid complex and medial hypothalamic nu- ectomy in rats with VMH or PVH lesions. Physiol Behav 45:
cleus functioning in food regulation. Physiol Behav 3: 567–570, 1131–1137, 1989.
1968. 70. Tokunaga K, Matsuzawa Y, Fujioka S, Kobatake T, Keno
60. Schoenfeld TA and Hamilton LW. Disruption of appetite but Y, Odaka H, Matsuo T, and Tarui S. PVN-lesioned obese rats
not hunger or satiety following small lesions in the amygdala of maintain ambulatory activity and its circadian rhythm. Brain
rats. J Comp Physiol Psychol 95: 565–587, 1981. Res Bull 26: 393–396, 1991.
61. Schwartz NB and Kling A. The effect of amygdaloid lesions on 71. White NM and Fisher AE. Relationship between amygdala
feeding, grooming and reproduction in rats. Acta Neuroveget 26: and hypothalamus in the control of eating behavior. Physiol
12–33, 1964. Behav 4: 199–205, 1969.
62. Schwartzbaum JS. Some characteristics of “amygdaloid hy- 72. Wood CD. Behavioral changes following discrete lesions of tem-
perphagia” in monkeys. Am J Physiol 74: 252–259, 1961. poral lobe structures. Neurology 8: 215–220, 1958.You can also read
