- Email: [email protected]
Plasticity in the propagation of hippocampal stimulation-induced activity: a [14C]2-deoxyglucose mapping study
Brain Research, 520 (1990) 199-207
199
Elsevier BRES 15604
Plasticity in the propagation of hippocampal stimulation-induced activity: a [14C]2-deoxyglucose mapping study Kenneth A. Campbell Department of Psychology, University of Pennsylvania, Philadelphia, PA 19104 ( U.S.A.)
(Accepted 12 December 1989) Key words: Hippocampus; Neuroplasticity; Kindling; Deoxyglucose; Metabolism; Self-stimulation
Previous work has suggested that the rewarding effect of hippocampal stimulation develops only as a consequence of neuroplastic change resulting from repeated stimulation experience: naive rats will not initiate self-stimulation for several days, but a prior program of hippocampal kindling greatly facilitates acquisition of self-stimulation. In the current study, the metabolic tracer [14C]2-deoxyglucosewas used to measure functional activity during a session of repeated electrical stimulation of the hippocampus to determine whether the resulting activation propagates more widely in experienced hippocampal self-stimulators than in naive rats receiving the same programmed stimulation. A control group of naive unstimulated rats was also included. The results indicated that dorsal CA3 stimulation in naive rats failed to increase metabolic activity in the hippocampus, while in experienced rats, the stimulation produced significant bilateral activation of CA1, CA3, and the ventral subiculum. These results provide support for the hypothesis that development of the rewarding effect of hippocampal stimulation is associated with more widespread propagation of stimulation-induced activity. INTRODUCTION Rats typically acquire lever-pressing tasks very rapidly when either a natural reward or medial forebrain bundle stimulation is used as the reinforcer 9'17. In marked contrast, rats are very slow to acquire the lever-pressing response for hippocampal stimulation (8-14 daily 30-min sessions), yet become reliable self-stimulators 16. Acquisition of hippocampal self-stimulation can be facilitated (to 1-3 sessions) by a prior program of repeated, daily electrical stimulation (kindling) applied to the hippocampal electrode 6. This stimulation pretreatment is effective even though it is administered in the absence of the situational and contingency cues of the lever-pressing chamber. Thus, the facilitative effect of pretreatment appears to depend on some change in the neural effects of stimulation per se and not on behavioral conditioning. In fact, the facilitation of acquisition appears to depend upon a neuroplastic property of the kindling stimulation such as potentiation: pretreatment with an equivalent number of pulses at a non-potentiating frequency (1 Hz) is ineffective in facilitating acquisition 4. An explanatory hypothesis has been advanced that the development of the rewarding effect of hippocampal stimulation may depend on potentiation of neural transmission between the hippocampus and some neural reinforcement system 5'6. According to this hypothesis,
acquisition of hippocampal self-stimulation is retarded in naive rats because the hippocampal stimulation is not rewarding, due to insufficient propagation of stimulationinduced activity. Emergence of hippocampal reward would then require trans-synaptic potentiation to allow stimulation-induced activity to propagate more widely, eventually activating some system capable of motivating behavior. The issue of reward as the psychological factor affected by stimulation pretreatment has been addressed in previous papers 3'4'7. Studies of the neural mechanism of the effect have demonstrated that the change is not restricted to the locally stimulated site, but is transferable to distant sites: stimulation pretreatment applied to one hippocampus is also effective in facilitating acquisition of self-stimulation tested in the contralateral hippocampus 5. Electrophysiological evidence indicates that this type of transfer of kindling effects probably involves transsynaptic potentiation mechanisms 8"19"2°. Thus, the positive transfer of pretreatment facilitation of acquisition to the contralateral hippocampus is consistent with the hypothesis that trans-synaptic potentiation mechanisms allow increased propagation of stimulation-induced activity. The present experiment was designed to provide an independent test of the hypothesis that the neural activation resulting from hippocampal stimulation propagates more widely in experienced, self-stimulating rats
* Correspondence: K.A. Campbell. Present address: Department of Psychology, University of Delaware, Newark, DE 19716, U.S.A.
0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
200 than
in n a i v e
rats
that
will
not
self-stimulate.
The
[14C]2-deoxyglucose (2-DG) autoradiographic technique was used to measure limbic metabolic activity resulting from programmed
hippocampal
stimulation, comparing
experienced hippocampal self-stimulators to stimulationnaive rats, and to unstimulated controls.
MATERIALS AND METHODS
Subjects and surgery Subjects were 19 male albino Sprague-Dawley rats (Charles River) weighing 339-584 g at surgery and 400-662 g at the time of 2-DG injection and sacrifice. Under pentobarbital anesthesia (60 mg/kg i.p.), all animals were implanted with monopolar stainless steel electrodes (0.25 mm bare; Formvar insulated) into the CA3 region of the dorsolateral hippocampus. Coordinates used were 3.8 mm posterior to bregma, 4.2 mm lateral to the midline, and 4.3 mm ventral to the level skull surface at bregma.
Apparatus and procedure Animals were tested in a Plexiglas Skinner box (26 cm each side × 46 cm high) with a hardware cloth floor. Stimulation trains were provided from a constant-current stimulator and consisted of 0.5 s of 80/~A, 0.1 ms rectangular monophasic-negative pulses at 75/s. These parameters were found to be optimal for supporting hippocampal self-stimulation, while minimizing seizure activity. Prior to the day of 2-DG injection, 7 animals were trained to respond for hippocampal stimulation until stable rates were obtained. All of these animals were-given a self-stimulation session on the day preceding 2-DG injection. The remaining animals were habituated to the chamber for 30-50 min sessions over 2-20 days. Animals were deprived of food for 16-20 h prior to 2-DG injection. Injection of 30/~Ci of [14C]2-DG in saline (i.p.) was given at the beginning of the 45 min 2-DG uptake period. During this period, 5 confirmed self-stimulating rats received hippocampal stimulation: two rats were allowed to self-stimulate for 45 min (total trains 352 and 449, respectively), while the other 3 rats were given programmed stimulation at 1 train every 5 s for a total of 540-592 trains. Stimulation-naive rats were either given the same programmed stimulation (n = 7) or were not stimulated (n = 5). In addition, two more confirmed self-stimulating rats were not stimulated during 2-DG uptake. Electrographic recordings were taken from the stimulating electrode during the interstimulus interval to monitor possible epileptiform activity during 2-DG uptake. At the end of the 45 min 2-DG uptake period, the animals were sacrificed with an overdose of Chloropent and perfused transcardially with a 3.3% buffered formalin solution. Brains were removed and frozen in Freon-22, and subsequently frozen sectioned at 20/~m and dried on a slide-warming tray. For autoradiography, sections were apposed to Kodak SB-5 X-ray film for 10-20 days to obtain a median optical density between 0.5 and 0.9. Sections were then stained with thionin.
coronal section was measured by computing the mean ROD of all image pixels within the borders of that structure. The computerized image analysis system permits delineation of the brain structure on the basis of the histological section, after it is aligned with the corresponding stored autoradiographic image. For the present study, 4 coronal levels of the hippocampal formation were chosen, as shown in Fig. 1. Detailed quantitative image analysis was performed at each of 3 levels (1, 3 and 4), excluding level 2 through the electrode locus, because the experimental hypothesis addressed propagated neural activity. The anatomical subdivisions defined in Fig. 1 were chosen for quantification based on known connectivity with the hippocampal CA3 stimulation site. Each structure was outlined by moving a cursor over the aligned histological image according to drawing rules also illustrated in Fig. i. The average optical density of the region so outlined was computed from the corresponding stored autoradiographic image to produce the mean ROD measure employed. Thus, at each of 3 coronal levels of the hippocampus, 5-7 structures were quantified on each side of the brain in two different sections in each animal.
Statistical treatment An analysis of variance (ANOVA) was performed for each of 3 coronal levels with two within-subjects factors, brain structure and side of the brain, and between-subjects factors, treatment groups and subjects-within-groups. Since the R O D is a proportion and thus not normally distributed at values approaching 100, the arcsin transformation was used. The analysis was performed on a VAX 730 computer (Digital Equipment Corporation, Maynard, MA) using the SAS generalized linear models procedure for multifactor ANOVA (SAS Institute, Cary, NC), with planned comparisons between groups for each structure performed by F-tests for specific contrasts. One family of comparisons was made at each coronal level. Within the structure x group interaction, pairwise comparisons between groups were made for bilateral effects in each structure, using the mean square deviation for structure x subjectswithin-groups as an error term. An error rate of P < 0.05 was adopted for each family of comparisons, and apportioned equally among individual comparisons, so that the significance level for comparisons at different coronal levels varied depending on the number of structures at each coronal level. The significance level per comparison so determined was P < 0.0024 for level 1 (3 x 7 comparisons per family), P < 0.0028 for level 3 (3 × 6) and P < 0.0033 for level 4 (3 × 5).
RESULTS
Histology E l e c t r o d e tip loci f o r all a n i m a l s t h a t r e c e i v e d s t i m u l a t i o n a r e i l l u s t r a t e d in Fig. l B . A l l e l e c t r o d e p l a c e m e n t s w e r e l o c a t e d in field C A 3 o f t h e d o r s o l a t e r a l h i p p o c a m p u s o r in t h e i m m e d i a t e l y a d j a c e n t f i m b r i a , a c r o s s t h e r e g i o n i d e n t i f i e d as s u p p o r t i n g h i p p o c a m p a l
Autoradiographic image analysis
l a t i o n in p r e v i o u s s t u d i e s 4. T h e r e
Autoradiographic images were analyzed on the computerized video densitometry system at the Biotechnology Resource Center at Drexel University, described in detail by Gallistel and collaborators TM. These authors also present the rationale for using a normalized index of metabolic activation, rather than estimates of glucose utilization, when the experimenter is interested in detecting discrete, localized changes in activity. Gallistel et al. 1° compared 3 different normalized indices of metabolic activity and found that the most robust and sensitive measure was the mean relative optical density (ROD). The ROD of a pixel (digitized 50/~m square spot on the image) is the percentile rank of its darkness compared to all the pixels in the brain section image. Thus, in the present study, the functional metabolic activity of a brain structure in a particular
d i f f e r e n c e s in e l e c t r o d e loci b e t w e e n g r o u p s .
were
self-stimu-
no consistent
Seizure activity Electrographic recordings made during 2-DG uptake in e x p e r i e n c e d a n i m a l s i n d i c a t e d t h a t in 3 o f t h e 5 c a s e s o n l y t h e first s t i m u l u s e l i c i t e d a n a f t e r d i s c h a r g e ( A D ) , in one
case
no
AD
was
elicited,
and
in o n e
case
no
r e c o r d i n g w a s m a d e . l n t e r i c t a l s p i k i n g w a s o b s e r v e d in one of these animals.
AD
was accompanied
f o r e l i m b c l o n u s in o n e e x p e r i e n c e d
by brief
animal while only
201
A. LEVEL 1
~
B. LEVEL 2
~
"DIG,'
,
2
Do LEVEL 4
C. LEVEL 3
/ Fig. 1. Anatomical divisions defined for quantitative image analysis at levels 1, 3 and 4 are labeled on the left sides of A, C and D, respectively, while shading on the right sides indicates significant differences between the experienced stimulated group and both naive groups. The direction of effect due to stimulation experience is denoted by arrows and distinctive shading. A: level 1 (-2.8 mm from bregma), anterior hippocampus and amygdala; B: level 2 (-3.8 mm), dorsal hippocampal electrode locus; C: level 3 (-5.8 mm), mid-hippocampal level; D: level 4 (-6.8 mm), posterior hippocampus. Symbols for electrode placements at level 2: naive-stimulated rats (dots), experienced-stimulated rats (solid triangles), and experienced-unstimulated rats (open triangles). Drawings are based on the atlas of Paxinos and Watson TM. AMYG, amygdaloid complex; lat, lateral half of the cortical amygdaloid complex; med, medial half of the cortical amygdaloid complex; bl, basolateral amygdaloid nucleus; CI, claustrum; DG, dentate gyrus; dSub, dorsal subiculum; HPC, hippocampus proper, subdivided at levels 3 and 4 into fields CA1 and CA3; pCg, posterior cingulate cortex; RSpl, retrospleniai cortex; vSub, ventral subiculum.
transient freezing was o b s e r v e d in the o t h e r 4 animals. Recordings m a d e in all 7 naive animals receiving prog r a m m e d stimulation indicated that no A D activity was elicited at the stimulating electrode: behavioral reactions to the stimulation included long arrests and, in one case, n u m e r o u s wet dog shakes.
Metabolic activity In the group of 5 confirmed self-stimulators that received stimulation during 2 - D G u p t a k e , no differences were found in the p a t t e r n of metabolic activation by c o m p a r i n g the two animals that were allowed to selfstimulate during 2 - D G u p t a k e to the 3 animals receiving p r o g r a m m e d stimulation, even though the self-stimulating animals received fewer stimulation trains. Thus,
the p r o g r a m m e d stimulation a p p e a r e d to be a satisfactory a p p r o x i m a t i o n to the p a t t e r n of trains received by self-administration, and u p t a k e in the regions e x a m i n e d a p p e a r e d not to be m a r k e d l y affected by o p e r a n t responding. Consequently, all 5 animals were p o o l e d t o g e t h e r to constitute the e x p e r i e n c e d - s t i m u l a t e d group, for comparison to the two naive groups. Level 1: anterior hippocampus. T h e r e were no significant differences b e t w e e n groups in the bilateral m e a n R O D s of the h i p p o c a m p u s , d e n t a t e gyrus, claustrum, or divisions of the a m y g d a l a at the coronal level 1.0 m m anterior to the stimulating electrode. D u e to the small cross-sectional area of the anterior h i p p o c a m p u s , no a t t e m p t was m a d e to quantify subdivisions of the hippocampus p r o p e r at this level. T h e only significant effect
202 TABLE I
Metabolic activity resuhingfrom CA3 stimulation Bilateral RODs (mean + S.E.M.) measured for each structure at each level illustrated in Fig. 1. See Fig. 1 for abbreviations of structure names.
Experienced Naive-Stim
Naive-Unstim
61.2 + 5.9 51.5 + 4.0 57.3 + 1.6 75.8 + 4.8* 49.7+2.0 33.5 + 0.6 22.5+2.2
57.3 + 4.3 53.0 + 4.2 57.3 + 1.7 89.9 + 2.2 49.0+2.1 40.5 + 4.2 26.0+3.1
Level 1 HPC DG Claustrum Post. cing. Amyg bl lat med
61.9 + 1.7 50.8 + 2.9 55.3 + 1.6 87.4 + 1.8 47.8+2.0 37.9 + 1.9 25.3+1.1
Level3 CA1 CA3 Dorsal DG DorsalSubic. Ventral Subic. Retrosplenial
71.9 + 5.2* 77.1 + 5.6* 50.0 + 2.5 59.0+4.5** 64.8 + 5.6* 68.1 + 3.2*
52.2 + 2.9 56.0 + 1.1 48.6 + 2.6 66.5+1.2"** 44.4 + 3.3 84.3 + 3.2
47.7 + 3.3 48.9 + 1.6 46.5 + 4.7 79.4+1.9 42.2 + 2.8 86.5 + 3.2
85.3 + 4.3* 50.4 + 4.3 76.5 + 3.3 76.2 + 6.8* 70.3 + 4.9
63.9 + 4.0 36.1 + 2.7 75.1 + 1.9 50.6 + 3.5 82.6 + 2.3
65.7 + 3.0 37.4 + 4.5 81.5 + 1.5 55.7 + 2.9 84.7 + 2.1
comparability of unilateral means suggests that the effects observed in the experienced animals are bilateral, despite the unilateral stimulation. In an A N O V A performed on only the unstimulated side for each group, the same pattern of results was obtained, except that the experienced versus naive-stimulated comparison for CA1 returned a P-level of 0.0033, thus not meeting the required P < 0.0028 level. There were no differences between groups in the dorsal dentate gyrus. In the case of the dorsal subiculum, each stimulated group was significantly suppressed compared to the unstimulated group. This was the only significant difference observed between the naive-stimulated and naive-unstimulated group at any level. Level 4: posterior hippocampus. At the posterior pole of the hippocampus depicted in Fig. 1D, 3.0 m m posterior to the stimulating electrode, area CA1 and the ventral subiculum showed significantly greater activity in the experienced animals than in either the naive-stimulated or naive-unstimulated groups, which again did not
Level 4 CA1 DG Dorsal Subic. Ventral Subic. Retrosplenial
*P < 0.0005, F-tests: experienced group vs each of the naive groups separately. **P < 0.0001, F-test: experienced group vs naive-unstimulated group; the comparison to the naive-stimulated group was not significant. ***P < 0.0024, F-test: naive-stimulated group vs naive-unstimulated group.
LEVEL m ¸
3
il II iI
m.
CA3
RETROSR.ENIAL
R.
O.
O.
D.
D.
IO
NU
0
E8
NU
CAt
observed at this level was a significant suppression of activity in the posterior cingulate cortex (Fig. 1A) in the experienced group compared to either the naive-stimulated group or the naive-unstimulated group, which did not differ (see Table I). This suppression appeared to be truly bilateral, as the m e a n R O D s for the stimulated side were within 1 S.E.M. of the mean of the unstimulated side of the brain in each group. Level 3: mid-hippocampus. The overall pattern of results at level 3, 2.0 mm posterior to the stimulating electrode, is illustrated in the right half of Fig. 1C. Experienced-stimulated animals showed significantly greater activity in hippocampal fields CA1 and CA3 and the ventral subiculum and significantly lower activity in the retrosplenial cortex than either the naive-stimulated or naive-unstimulated groups, which did not differ (see Table I). The magnitudes of these differences are illustrated by the bar graphs in Fig. 2, which also indicate the effects on the stimulated versus unstimulated side. Although the effects appear to be consistently stronger on the stimulated side in the hippocampus proper, the
v~n'RAL
R.
R.
O.
O.
D.
D.
NU
m
o
ES
LEVEL Ira-
O.
D.
D.
NS
8LmlCULUM
N8
E8
~JBIOJI.UM
1
R. O.
NU
E8
4 laa
CA1
NU
m
o ES
Fig. 2. Mean ROD (+ S.E.M.) for each group for structures at levels 3 and 4 that showed significant differences in pairwise comparisons between the experienced-stimulated group (ES) and both the naive-stimulated (NS) group, and the naive-unstimulated group (NU). Implanted/stimulated side, hatched bars; unstimulated side, open bars.
203
Level 3
____fruit f~,'x_%\_.
~T_~ ~ { r--~. "~ .,,..-,~<~::~,~-.~ ~, ..~
Experienced-Stimulated
Naive.Stimulated
Naive-Unstimulated
,..~?'=2- k=..~ ,,
((
/ •
/
fJ~-..
.~
,..-.,-.-.
x'~ i,, ~ )
.!ji,/
:)J!'
(.t.~_......._)_,_,~_.,~_.....__.
,f
..... ~
:)))) " ....
-,..
,'
~I
----.
/F,-"--r"-
-"'"c-"'-.~
",,,
It," i
\,,,~.,i/'---
-f'-''
,i
.~
..~
"
•,-- . . . . . ~ 1
.,_~.~ _ ~ , _ . ~ ' ,f--~'~,-~
• .'-..; ';..(',",. ,
I ("l ........... ,. . . . .
. .., .,_t.
_____. ..
i - . . . . . . . ~ I "-~'~-~'--
,--.~.,~-m J( ~ . . . \
I-'.L__./.,'~ '.'
..... ...--~___..--',-.: ..... ..,
LI.& i___ ,."
, .q,~.,.-
"\
...~
t,ti k.
\,,,t 1
ilL
.~S~/,,/'v~.~-~-,'c.-, ".
/'r~" $,/~ IL'~= ' h*"~;II': , t!
k
\,. .... .--Y-,. :.-/
/'lJ ~ ,,/
Fig. 3. Drawings of a mid-hippocampal section at level 3 from each of 5 animals in each group, showing the pixels in the hippocampal region that fell within an optical density window of the darkest (most active) 7-10% of pixels in the image. The stimulated side is left, except for the last section in the experienced group and the last two sections in the naive group.
differ (Table I). As shown in the bar graphs of Fig. 2, these results parallel those in the same structures at level 3. In addition, it is apparent from the unilateral means depicted in Fig. 2 that both CA1 and the ventral subiculum were comparably affected on the stimulated and unstimulated side. Again, an A N O V A performed on
the data from only the unstimulated side yielded the same pattern of significant results. There were no differences between groups in the dentate gyrus or the dorsal subiculum. This latter finding fails to support the suppression of the dorsal subiculum observed in both stimulated groups at level 3. T h e
204
Level 4 Naive-Unstimulated •
---
"
.1-?
---"i;
.. ..... T,-~f-~_:.-_~.. ,, ,, ,,/i "/. "'i: \\ '\
.................
~<;,.. ~'~..
i ,//,~,
,
,......,f--..--.J _~>. ~,
"~x\~
,, k
,:" ..~__~,,,~-.-..~. ,.
l/;:ir
,
i I(.~(
\"
-.
I'o~" '~
~ .
..... /
.,
~ x
t
Ig-~
I/
--~'~.- \
' /
r'
.,~,
~ i
",.~_ ~ _ ~
",____~.."
'",
~".
~
.,,",.~< /~/ r:~-//
........ 4b" ........ \..,YS~,-" ~-~.,\
~-~-~-~
-;
\
I
I
- ~_~
.... ~ . . . . . . .
ii
.1?~"'-*k _="
Y'/? I
....-.-"--~ .,.~:> .... IL ~"",, ,~- ,,'
-~...,,'.. '~,~ tl I
7, ...._.......( / +
Ii , i'~
i
._.__
Experienced.Stimulated
Naive-Stimulated
I
~J
--/
'\
Lb.:~'k
.,c.C_.~, /
¢
..
lj ''~ ,.,& \., -., \:~ \
\
/i
j, f
_
j,F~,
17
I ......
)' I,
~
-
-
",\..
i ././J ~
,3,...
°
~\',, ./.
J
Fig. 4. Drawings of a posterior hippocampal section at level 4 from each of 5 animals in each group, showing the pixels in the hippocampal region that fell within an optical density window of the darkest 7-10% of pixels in the image. The relative positions of brain sections correspond exactly to those in Fig. 3,
significant suppression in the retrosplenial cortex observed in the experienced animals at level 3 did not achieve significance in the corresponding measurements at level 4, although the means did show a trend to suppression (Table I). Foci o f activation To examine the anatomical localization of activation within the hippocampus and its variability within and
between groups, an optical density window routine was employed l°'n with the results shown in Figs. 3 and 4. After major landmarks were drawn, the window procedure produced a stippling that corresponds to the range of R O D values within the window. In this case the window was used to highlight the darkest 7-10% of pixels in the image (pixels with R O D values above the 90-93rd percentile), and then display those found in the hippocampai region. The larger proportion of darkened pixels
205 in the hippocampi of experienced-stimulated animals in Figs. 3 and 4 presents a clear anatomical demonstration of the significant quantitative effects reported above. As expected, the greatest activation appears in areas CA1 and CA3, with the dorsal-ventral extent of activation being broad and variable. Effects also appear to be more intense on the stimulated side (which is left, in the 4 experienced animals showing the largest effects). Also apparent is the overall absence of effect of stimulation in naive animals: in most cases, the naive-stimulated sections are indistinguishable from naive-unstimulated sections. Individual differences within groups may provide additional information related to the behavioral effects of stimulation. The last animal in the experienced-stimulated group shows considerably less activation compared to other animals in the same group at both levels 3 and 4, while the last animal in the naive-stimulated group shows relatively intense activation in CA1 at level 4 (see Figs. 3 and 4). Previous studies demonstrating that hippocampal stimulation pretreatment facilitates acquisition of hippocampal self-stimulation have frequently shown exceptional animals in the naive group that learned rapidly, unlike their cohorts 4'6. These exceptional animals may share common factors with the fifth animal shown in the naive-stimulated group in Fig. 4, such that stimulation results in wide propagation even in a naive animal. It has not been determined whether such factors might be extrinsic (such as electrode placement or vascular damage during implantation) or intrinsic (such as biochemical differences interacting with recruitment). In the two confirmed self-stimulators that were not stimulated during 2-DG uptake, measurements of areas CA1 and CA3 at levels 3 and 4 yielded mean ROD values that were within 1 S.E.M. of the means for the naive-unstimulated group given in Table I. This result is consistent with earlier studies in which rats previously kindled to behavioral convulsions showed no effect on 2-DG uptake in the absence of stimulation of the kindled amygdala 2 or the kindled hippocampus 15. Thus, the significant elevation of metabolic activity in the experienced-stimulated group can be attributed to the propagation of activity triggered by hippocampal stimulation administered during 2-DG uptake, and not to a long-term metabolic effect of previous stimulation experience alone. DISCUSSION The principal findings of this study indicate that hippocampal stimulation induces significantly more widespread propagation of activity in animals experienced with the stimulation through prior self-stimulation train-
ing than in naive animals. With the electrode loci and stimulation parameters employed in the present experiment, no evidence was found that over 500 stimulation trains delivered to the hippocampus in naive rats produced any significant metabolic activation of the structures examined at 1 mm anterior, or 2 or 3 mm posterior to the stimulating electrode. In contrast, experienced animals showed strong bilateral activation of the hippocampus proper and ventral subiculum at 2 and 3 mm posterior to the stimulating electrode. This result demonstrates that the 2-DG metabolic mapping technique is capable of revealing differential propagation due to use-induced plasticity even with modest stimulation parameters. The one previous report of increased 2-DG uptake in animals with prior hippocampal stimulation experience utilized a deliberately epileptogenic rapid kindling procedure 15. The present results provide experimental support for the neural correlate proposed in the hypothesis that hippocampal stimulation is not reinforcing in naive rats because of insufficient propagation, but acquires a reinforcing effect as stimulation experience engenders more widespread stimulation-induced activation. The results clearly indicate that even a large number of hippocampal stimulation trains in naive rats fail to propagate significantly within the hippocampus, while the same program of stimulation in experienced animals propagates widely throughout the hippocampus. While these findings support the contention of differential propagation as a function of experience, no direct inference can be made regarding the mechanism of this effect. The increased extent of metabolic labeling may be due to a potentiation-like mechanism, or to other neuroplastic effects of kindling. Studies demonstrating that stimulation pretreatment transfers its facilitative effect on acquisition of hippocampal self-stimulation to the contralateral hippocampus support trans-synaptic mechanisms over local excitability effects5. Possibly, multiple effects such as potentiation as well as decreased G A B A inhibition following hippocampal kindling 13 may contribute to the wider propagation of stimulationinduced activity observed in the hippocampus of experienced animals. Other effects that may contribute to the widespread propagation in experienced animals relate to the elicitation of epileptiform activity. Even though a low current level was chosen to minimize seizures, some seizure activity was elicited in experienced animals, but not observed in naive animals. Although AD was triggered by only the first stimulation train, it is possible that the widespread activation of the hippocampus observed in experienced animals is at least partly due to the initial epileptiform activity. Consequently, in the present ex-
206 periment, the naive and experienced groups differed not only in the presumed rewarding effect of the stimulation, but also in elicited A D activity. Thus, one cannot rule out the possibility that at least part of the increase in propagation observed in experienced animals is a correlate of hippocampal seizure rather than of hippocampal reward. Hippocampal reward need not be accompanied by epileptiform activity: previous studies have shown that maintenance of hippocampal self-stimulation does not depend on elicitation of A D activity 5. The pattern of activation resulting from hippocampal stimulation trains in experienced animals observed in the present experiment is in general agreement with previous studies of metabolic labeling during dorsal hippocampal stimulation. The most consistent finding is that unilateral hippocampal stimulation produces strong bilateral activation of hippocampal fields CA1 and CA314"15'23: This bilateral activation is consistent with the extensive commissural connections of the hippocampus, and with the electrode placement in CA3, from which the commissural fibers originate that project to contralateral CA1, CA3, and subiculum 22. Due to the more intense stimulation treatments administered in previous studies, extra-hippocampal activation reported in areas such as the amygdala 14"15 was not observed in the present experiment. However, the significant activation of the ventral subiculum at both levels 3 and 4 in the present experiment has been noted in only a few previous cases 14. There have been no previous reports of decreases in metabolic labeling resulting from stimulation of the hippocampus, but in the present experiment, an unex-
REFERENCES 1 Ackermann, R.E, Finch, D.M., Babb, T.L. and Engel Jr., J., Increased glucose metabolism during long-duration recurrent inhibition of hippocampal pyramidal cells, J. Neurosci., 4 (1984) 251-264. 2 Biackwood, D.H.R., Kapoor, V. and Martin, M.J., Regional changes in cerebral glucose utilization associated with amygdaloid kindling and electroshock in the rat, Brain Research, 224 (1981) 204-208. 3 Campbell, K.A., Development of the reinforcing effects of hippocampal stimulation through repeated daily electrical stimulation (kindling) (Doctoral dissertation, University of Toronto, 1982), Dissert. Abs. Internat., 43 (1983) 4189B. 4 Campbell, K.A. and Milgram, N.W., Potentiating hippocampal stimulation facilitates acquisition of lever-pressing for stimulation but not food, Physiol. Behav., 24 (1980) 1115-1118. 5 Campbell, K.A. and Milgram, N.W., Mechanisms underlying the plasticity of hippocampal stimulation-induced reward, Behay. Neurosci., 99 (1985) 209-219. 6 Campbell, K.A., Milgram, N.W. and Christoff, J.K., Plasticity in the reinforcing consequences of hippocampal stimulation, Brain Research, 159 (1978) 458-462. 7 Caudarella, M., Campbell, K.A. and Milgram, N.W., Hippocampal stimulation limits performance but does not retard acquisition of food-reinforced learning, Physiol. Behav., 32 (1984) 895-898.
pected consistent bilateral suppression of metabolic activity was observed in the posterior cingulate and retrosplenial cortex of experienced animals relative to the naive groups. This difference was observed at levels 1, 3 and 4, although the comparison was significant at levels 1 and 3 only. Because the R O D expresses metabolic activity of a structure relative to the image as a whole, it is conceivable that the suppression observed in the cingulate-retrosplenial cortex might be an artifact of the increase in activity over relatively large areas of the hippocampus in the experienced group. Observations mitigating against this interpretation are the fact that the suppression was significant at level 1 where no increases in metabolism were seen, and the lack of effect on other structures such as the dentate gyrus at levels 3 and 4, where large areas of activation were seen. Since CA3 is known to project to the posterior cingulate area 21, the suppression observed in the cingulate region in the experienced group may reflect inhibitory activity propagated as a result of stimulation. However, combined electrophysiologic and deoxyglucose labeling studies in the hippocampus have indicated that the metabolic demands of inhibitory interneurons can render paradoxical correlations between metabolic activity and excitation-inhibition of principal neurons 1. Acknowledgements. This work was supported by National Science Foundation Grant BNS 82 11972 to C.R. Gallistel; image processing was supported by National Institutes of Health Biotechnology Resource Center Grant 23-1352630. The author was supported on a postdoctoral fellowship from National Institute for Mental Health Training Grant T32 MH15092. The author would like to thank Elysa Braunstein for technical assistance.
8 Douglas, R.M. and Goddard, G.V., Long-term potentiation of the perforant path-granule cell synapse in the rat hippocampus, Brain Research, 86 (1975) 205-215. 9 Estes, W.K., The statistical approach to learning theory. In S. Koch (Ed.), Psychology: A Study of a Science, Vol. 2, McGrawHill, New York, 1959, pp. 380-491. 10 Gallistel, C.R., Gomita, Y., Yadin, E. and Campbell, K.A., Forebrain origins and terminations of the MFB metabolically activated by rewarding stimulation or by reward-blocking doses of pimozide, J. Neurosci., 5 (1985) 1246-1261. 11 Gallistel, C.R., Piner, C.T., Allen, T.O., Adler, N.T., Yadin, E. and Negin, M., Computer assisted analysis of 2-DG autoradiographs, Neurosci. Biobehav. Rev., 6 (1982) 409-420. 12 Gallistel, C.R. and Tretiak, O., Microcomputer systems for analyzing 2-deoxyglucose autoradiographs. In R.R. Mize (Ed.), The Microcomputer in Cell and Neurobiology Research, Elsevier, New York, 1985, pp. 390-408. 13 Kapur, J., Stringer, J.L. and Lothman, E.W., Evidence that repetitive seizures in the hippocampus cause a lasting reduction of GABAergic inhibition, J. Neurophysiol., 61 (1989) 417-426. 14 Kliot, M. and Poletti, C.E., Hippocampal afterdischarges: differential spread of activity shown by the [14C]deoxyglucose technique, Science, 204 (1979) 641-643. 15 Lothman, E.W., Hatlelid, J.M. and Zorumski, C.E, Functional mapping of limbic seizures originating in the hippocampus: a combined 2-deoxyglucose and electrophysiologic study, Brain Research, 360 (1985) 92-100.
207 16 Milgram, N.W., Server, A.C. and Campbell, K.A., Effect of food and water deprivation on hippocampal self-stimulation and post-stimulation feeding, Physiol. Psychol., 5 (1977) 43-48. 17 Olds, M.E. and Fobes, J.L., The central basis of motivation: intracranial self-stimulation studies, Ann. Rev. Psychol., 32 (1981) 523-574. 18 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic, New York, 1982, 162 pp. 19 Racine, R., Gartner, J.G. and Burnham, W.M., Epileptiform activity and neural plasticity in brain structures, Brain Research, 47 (1972) 262-268. 20 Racine, R., Newberry, E and Burnham, W.M., Post-activation potentiation and the kindling phenomenon, Electroencephalogr.
Clin. Neurophysiol., 39 (1975) 261-271. 21 Swanson, L.W. and Cowan, W.M., An autoradiographic study of the organization of the efferent connections of the hippocampal formation in the rat, J. Comp. Neurol., 172 (1977) 49-84. 22 Swanson, L.W., Sawchenko, P.E. and Cowan, W.M., Evidence for collateral projections by neurons in Ammon's horn, the dentate gyrus, and the subiculum: a multiple retrograde labeling study in the rat, J. Neurosci., 1 (1981) 548-559. 23 Watson Jr., R.E., Edinger, H.M. and Siegel, A., A p4C]2deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. II. The hippocampal formation, Brain Res. Rev., 5 (1983) 133-176.
199
Elsevier BRES 15604
Plasticity in the propagation of hippocampal stimulation-induced activity: a [14C]2-deoxyglucose mapping study Kenneth A. Campbell Department of Psychology, University of Pennsylvania, Philadelphia, PA 19104 ( U.S.A.)
(Accepted 12 December 1989) Key words: Hippocampus; Neuroplasticity; Kindling; Deoxyglucose; Metabolism; Self-stimulation
Previous work has suggested that the rewarding effect of hippocampal stimulation develops only as a consequence of neuroplastic change resulting from repeated stimulation experience: naive rats will not initiate self-stimulation for several days, but a prior program of hippocampal kindling greatly facilitates acquisition of self-stimulation. In the current study, the metabolic tracer [14C]2-deoxyglucosewas used to measure functional activity during a session of repeated electrical stimulation of the hippocampus to determine whether the resulting activation propagates more widely in experienced hippocampal self-stimulators than in naive rats receiving the same programmed stimulation. A control group of naive unstimulated rats was also included. The results indicated that dorsal CA3 stimulation in naive rats failed to increase metabolic activity in the hippocampus, while in experienced rats, the stimulation produced significant bilateral activation of CA1, CA3, and the ventral subiculum. These results provide support for the hypothesis that development of the rewarding effect of hippocampal stimulation is associated with more widespread propagation of stimulation-induced activity. INTRODUCTION Rats typically acquire lever-pressing tasks very rapidly when either a natural reward or medial forebrain bundle stimulation is used as the reinforcer 9'17. In marked contrast, rats are very slow to acquire the lever-pressing response for hippocampal stimulation (8-14 daily 30-min sessions), yet become reliable self-stimulators 16. Acquisition of hippocampal self-stimulation can be facilitated (to 1-3 sessions) by a prior program of repeated, daily electrical stimulation (kindling) applied to the hippocampal electrode 6. This stimulation pretreatment is effective even though it is administered in the absence of the situational and contingency cues of the lever-pressing chamber. Thus, the facilitative effect of pretreatment appears to depend on some change in the neural effects of stimulation per se and not on behavioral conditioning. In fact, the facilitation of acquisition appears to depend upon a neuroplastic property of the kindling stimulation such as potentiation: pretreatment with an equivalent number of pulses at a non-potentiating frequency (1 Hz) is ineffective in facilitating acquisition 4. An explanatory hypothesis has been advanced that the development of the rewarding effect of hippocampal stimulation may depend on potentiation of neural transmission between the hippocampus and some neural reinforcement system 5'6. According to this hypothesis,
acquisition of hippocampal self-stimulation is retarded in naive rats because the hippocampal stimulation is not rewarding, due to insufficient propagation of stimulationinduced activity. Emergence of hippocampal reward would then require trans-synaptic potentiation to allow stimulation-induced activity to propagate more widely, eventually activating some system capable of motivating behavior. The issue of reward as the psychological factor affected by stimulation pretreatment has been addressed in previous papers 3'4'7. Studies of the neural mechanism of the effect have demonstrated that the change is not restricted to the locally stimulated site, but is transferable to distant sites: stimulation pretreatment applied to one hippocampus is also effective in facilitating acquisition of self-stimulation tested in the contralateral hippocampus 5. Electrophysiological evidence indicates that this type of transfer of kindling effects probably involves transsynaptic potentiation mechanisms 8"19"2°. Thus, the positive transfer of pretreatment facilitation of acquisition to the contralateral hippocampus is consistent with the hypothesis that trans-synaptic potentiation mechanisms allow increased propagation of stimulation-induced activity. The present experiment was designed to provide an independent test of the hypothesis that the neural activation resulting from hippocampal stimulation propagates more widely in experienced, self-stimulating rats
* Correspondence: K.A. Campbell. Present address: Department of Psychology, University of Delaware, Newark, DE 19716, U.S.A.
0006-8993/90/$03.50 (~) 1990 Elsevier Science Publishers B.V. (Biomedical Division)
200 than
in n a i v e
rats
that
will
not
self-stimulate.
The
[14C]2-deoxyglucose (2-DG) autoradiographic technique was used to measure limbic metabolic activity resulting from programmed
hippocampal
stimulation, comparing
experienced hippocampal self-stimulators to stimulationnaive rats, and to unstimulated controls.
MATERIALS AND METHODS
Subjects and surgery Subjects were 19 male albino Sprague-Dawley rats (Charles River) weighing 339-584 g at surgery and 400-662 g at the time of 2-DG injection and sacrifice. Under pentobarbital anesthesia (60 mg/kg i.p.), all animals were implanted with monopolar stainless steel electrodes (0.25 mm bare; Formvar insulated) into the CA3 region of the dorsolateral hippocampus. Coordinates used were 3.8 mm posterior to bregma, 4.2 mm lateral to the midline, and 4.3 mm ventral to the level skull surface at bregma.
Apparatus and procedure Animals were tested in a Plexiglas Skinner box (26 cm each side × 46 cm high) with a hardware cloth floor. Stimulation trains were provided from a constant-current stimulator and consisted of 0.5 s of 80/~A, 0.1 ms rectangular monophasic-negative pulses at 75/s. These parameters were found to be optimal for supporting hippocampal self-stimulation, while minimizing seizure activity. Prior to the day of 2-DG injection, 7 animals were trained to respond for hippocampal stimulation until stable rates were obtained. All of these animals were-given a self-stimulation session on the day preceding 2-DG injection. The remaining animals were habituated to the chamber for 30-50 min sessions over 2-20 days. Animals were deprived of food for 16-20 h prior to 2-DG injection. Injection of 30/~Ci of [14C]2-DG in saline (i.p.) was given at the beginning of the 45 min 2-DG uptake period. During this period, 5 confirmed self-stimulating rats received hippocampal stimulation: two rats were allowed to self-stimulate for 45 min (total trains 352 and 449, respectively), while the other 3 rats were given programmed stimulation at 1 train every 5 s for a total of 540-592 trains. Stimulation-naive rats were either given the same programmed stimulation (n = 7) or were not stimulated (n = 5). In addition, two more confirmed self-stimulating rats were not stimulated during 2-DG uptake. Electrographic recordings were taken from the stimulating electrode during the interstimulus interval to monitor possible epileptiform activity during 2-DG uptake. At the end of the 45 min 2-DG uptake period, the animals were sacrificed with an overdose of Chloropent and perfused transcardially with a 3.3% buffered formalin solution. Brains were removed and frozen in Freon-22, and subsequently frozen sectioned at 20/~m and dried on a slide-warming tray. For autoradiography, sections were apposed to Kodak SB-5 X-ray film for 10-20 days to obtain a median optical density between 0.5 and 0.9. Sections were then stained with thionin.
coronal section was measured by computing the mean ROD of all image pixels within the borders of that structure. The computerized image analysis system permits delineation of the brain structure on the basis of the histological section, after it is aligned with the corresponding stored autoradiographic image. For the present study, 4 coronal levels of the hippocampal formation were chosen, as shown in Fig. 1. Detailed quantitative image analysis was performed at each of 3 levels (1, 3 and 4), excluding level 2 through the electrode locus, because the experimental hypothesis addressed propagated neural activity. The anatomical subdivisions defined in Fig. 1 were chosen for quantification based on known connectivity with the hippocampal CA3 stimulation site. Each structure was outlined by moving a cursor over the aligned histological image according to drawing rules also illustrated in Fig. i. The average optical density of the region so outlined was computed from the corresponding stored autoradiographic image to produce the mean ROD measure employed. Thus, at each of 3 coronal levels of the hippocampus, 5-7 structures were quantified on each side of the brain in two different sections in each animal.
Statistical treatment An analysis of variance (ANOVA) was performed for each of 3 coronal levels with two within-subjects factors, brain structure and side of the brain, and between-subjects factors, treatment groups and subjects-within-groups. Since the R O D is a proportion and thus not normally distributed at values approaching 100, the arcsin transformation was used. The analysis was performed on a VAX 730 computer (Digital Equipment Corporation, Maynard, MA) using the SAS generalized linear models procedure for multifactor ANOVA (SAS Institute, Cary, NC), with planned comparisons between groups for each structure performed by F-tests for specific contrasts. One family of comparisons was made at each coronal level. Within the structure x group interaction, pairwise comparisons between groups were made for bilateral effects in each structure, using the mean square deviation for structure x subjectswithin-groups as an error term. An error rate of P < 0.05 was adopted for each family of comparisons, and apportioned equally among individual comparisons, so that the significance level for comparisons at different coronal levels varied depending on the number of structures at each coronal level. The significance level per comparison so determined was P < 0.0024 for level 1 (3 x 7 comparisons per family), P < 0.0028 for level 3 (3 × 6) and P < 0.0033 for level 4 (3 × 5).
RESULTS
Histology E l e c t r o d e tip loci f o r all a n i m a l s t h a t r e c e i v e d s t i m u l a t i o n a r e i l l u s t r a t e d in Fig. l B . A l l e l e c t r o d e p l a c e m e n t s w e r e l o c a t e d in field C A 3 o f t h e d o r s o l a t e r a l h i p p o c a m p u s o r in t h e i m m e d i a t e l y a d j a c e n t f i m b r i a , a c r o s s t h e r e g i o n i d e n t i f i e d as s u p p o r t i n g h i p p o c a m p a l
Autoradiographic image analysis
l a t i o n in p r e v i o u s s t u d i e s 4. T h e r e
Autoradiographic images were analyzed on the computerized video densitometry system at the Biotechnology Resource Center at Drexel University, described in detail by Gallistel and collaborators TM. These authors also present the rationale for using a normalized index of metabolic activation, rather than estimates of glucose utilization, when the experimenter is interested in detecting discrete, localized changes in activity. Gallistel et al. 1° compared 3 different normalized indices of metabolic activity and found that the most robust and sensitive measure was the mean relative optical density (ROD). The ROD of a pixel (digitized 50/~m square spot on the image) is the percentile rank of its darkness compared to all the pixels in the brain section image. Thus, in the present study, the functional metabolic activity of a brain structure in a particular
d i f f e r e n c e s in e l e c t r o d e loci b e t w e e n g r o u p s .
were
self-stimu-
no consistent
Seizure activity Electrographic recordings made during 2-DG uptake in e x p e r i e n c e d a n i m a l s i n d i c a t e d t h a t in 3 o f t h e 5 c a s e s o n l y t h e first s t i m u l u s e l i c i t e d a n a f t e r d i s c h a r g e ( A D ) , in one
case
no
AD
was
elicited,
and
in o n e
case
no
r e c o r d i n g w a s m a d e . l n t e r i c t a l s p i k i n g w a s o b s e r v e d in one of these animals.
AD
was accompanied
f o r e l i m b c l o n u s in o n e e x p e r i e n c e d
by brief
animal while only
201
A. LEVEL 1
~
B. LEVEL 2
~
"DIG,'
,
2
Do LEVEL 4
C. LEVEL 3
/ Fig. 1. Anatomical divisions defined for quantitative image analysis at levels 1, 3 and 4 are labeled on the left sides of A, C and D, respectively, while shading on the right sides indicates significant differences between the experienced stimulated group and both naive groups. The direction of effect due to stimulation experience is denoted by arrows and distinctive shading. A: level 1 (-2.8 mm from bregma), anterior hippocampus and amygdala; B: level 2 (-3.8 mm), dorsal hippocampal electrode locus; C: level 3 (-5.8 mm), mid-hippocampal level; D: level 4 (-6.8 mm), posterior hippocampus. Symbols for electrode placements at level 2: naive-stimulated rats (dots), experienced-stimulated rats (solid triangles), and experienced-unstimulated rats (open triangles). Drawings are based on the atlas of Paxinos and Watson TM. AMYG, amygdaloid complex; lat, lateral half of the cortical amygdaloid complex; med, medial half of the cortical amygdaloid complex; bl, basolateral amygdaloid nucleus; CI, claustrum; DG, dentate gyrus; dSub, dorsal subiculum; HPC, hippocampus proper, subdivided at levels 3 and 4 into fields CA1 and CA3; pCg, posterior cingulate cortex; RSpl, retrospleniai cortex; vSub, ventral subiculum.
transient freezing was o b s e r v e d in the o t h e r 4 animals. Recordings m a d e in all 7 naive animals receiving prog r a m m e d stimulation indicated that no A D activity was elicited at the stimulating electrode: behavioral reactions to the stimulation included long arrests and, in one case, n u m e r o u s wet dog shakes.
Metabolic activity In the group of 5 confirmed self-stimulators that received stimulation during 2 - D G u p t a k e , no differences were found in the p a t t e r n of metabolic activation by c o m p a r i n g the two animals that were allowed to selfstimulate during 2 - D G u p t a k e to the 3 animals receiving p r o g r a m m e d stimulation, even though the self-stimulating animals received fewer stimulation trains. Thus,
the p r o g r a m m e d stimulation a p p e a r e d to be a satisfactory a p p r o x i m a t i o n to the p a t t e r n of trains received by self-administration, and u p t a k e in the regions e x a m i n e d a p p e a r e d not to be m a r k e d l y affected by o p e r a n t responding. Consequently, all 5 animals were p o o l e d t o g e t h e r to constitute the e x p e r i e n c e d - s t i m u l a t e d group, for comparison to the two naive groups. Level 1: anterior hippocampus. T h e r e were no significant differences b e t w e e n groups in the bilateral m e a n R O D s of the h i p p o c a m p u s , d e n t a t e gyrus, claustrum, or divisions of the a m y g d a l a at the coronal level 1.0 m m anterior to the stimulating electrode. D u e to the small cross-sectional area of the anterior h i p p o c a m p u s , no a t t e m p t was m a d e to quantify subdivisions of the hippocampus p r o p e r at this level. T h e only significant effect
202 TABLE I
Metabolic activity resuhingfrom CA3 stimulation Bilateral RODs (mean + S.E.M.) measured for each structure at each level illustrated in Fig. 1. See Fig. 1 for abbreviations of structure names.
Experienced Naive-Stim
Naive-Unstim
61.2 + 5.9 51.5 + 4.0 57.3 + 1.6 75.8 + 4.8* 49.7+2.0 33.5 + 0.6 22.5+2.2
57.3 + 4.3 53.0 + 4.2 57.3 + 1.7 89.9 + 2.2 49.0+2.1 40.5 + 4.2 26.0+3.1
Level 1 HPC DG Claustrum Post. cing. Amyg bl lat med
61.9 + 1.7 50.8 + 2.9 55.3 + 1.6 87.4 + 1.8 47.8+2.0 37.9 + 1.9 25.3+1.1
Level3 CA1 CA3 Dorsal DG DorsalSubic. Ventral Subic. Retrosplenial
71.9 + 5.2* 77.1 + 5.6* 50.0 + 2.5 59.0+4.5** 64.8 + 5.6* 68.1 + 3.2*
52.2 + 2.9 56.0 + 1.1 48.6 + 2.6 66.5+1.2"** 44.4 + 3.3 84.3 + 3.2
47.7 + 3.3 48.9 + 1.6 46.5 + 4.7 79.4+1.9 42.2 + 2.8 86.5 + 3.2
85.3 + 4.3* 50.4 + 4.3 76.5 + 3.3 76.2 + 6.8* 70.3 + 4.9
63.9 + 4.0 36.1 + 2.7 75.1 + 1.9 50.6 + 3.5 82.6 + 2.3
65.7 + 3.0 37.4 + 4.5 81.5 + 1.5 55.7 + 2.9 84.7 + 2.1
comparability of unilateral means suggests that the effects observed in the experienced animals are bilateral, despite the unilateral stimulation. In an A N O V A performed on only the unstimulated side for each group, the same pattern of results was obtained, except that the experienced versus naive-stimulated comparison for CA1 returned a P-level of 0.0033, thus not meeting the required P < 0.0028 level. There were no differences between groups in the dorsal dentate gyrus. In the case of the dorsal subiculum, each stimulated group was significantly suppressed compared to the unstimulated group. This was the only significant difference observed between the naive-stimulated and naive-unstimulated group at any level. Level 4: posterior hippocampus. At the posterior pole of the hippocampus depicted in Fig. 1D, 3.0 m m posterior to the stimulating electrode, area CA1 and the ventral subiculum showed significantly greater activity in the experienced animals than in either the naive-stimulated or naive-unstimulated groups, which again did not
Level 4 CA1 DG Dorsal Subic. Ventral Subic. Retrosplenial
*P < 0.0005, F-tests: experienced group vs each of the naive groups separately. **P < 0.0001, F-test: experienced group vs naive-unstimulated group; the comparison to the naive-stimulated group was not significant. ***P < 0.0024, F-test: naive-stimulated group vs naive-unstimulated group.
LEVEL m ¸
3
il II iI
m.
CA3
RETROSR.ENIAL
R.
O.
O.
D.
D.
IO
NU
0
E8
NU
CAt
observed at this level was a significant suppression of activity in the posterior cingulate cortex (Fig. 1A) in the experienced group compared to either the naive-stimulated group or the naive-unstimulated group, which did not differ (see Table I). This suppression appeared to be truly bilateral, as the m e a n R O D s for the stimulated side were within 1 S.E.M. of the mean of the unstimulated side of the brain in each group. Level 3: mid-hippocampus. The overall pattern of results at level 3, 2.0 mm posterior to the stimulating electrode, is illustrated in the right half of Fig. 1C. Experienced-stimulated animals showed significantly greater activity in hippocampal fields CA1 and CA3 and the ventral subiculum and significantly lower activity in the retrosplenial cortex than either the naive-stimulated or naive-unstimulated groups, which did not differ (see Table I). The magnitudes of these differences are illustrated by the bar graphs in Fig. 2, which also indicate the effects on the stimulated versus unstimulated side. Although the effects appear to be consistently stronger on the stimulated side in the hippocampus proper, the
v~n'RAL
R.
R.
O.
O.
D.
D.
NU
m
o
ES
LEVEL Ira-
O.
D.
D.
NS
8LmlCULUM
N8
E8
~JBIOJI.UM
1
R. O.
NU
E8
4 laa
CA1
NU
m
o ES
Fig. 2. Mean ROD (+ S.E.M.) for each group for structures at levels 3 and 4 that showed significant differences in pairwise comparisons between the experienced-stimulated group (ES) and both the naive-stimulated (NS) group, and the naive-unstimulated group (NU). Implanted/stimulated side, hatched bars; unstimulated side, open bars.
203
Level 3
____fruit f~,'x_%\_.
~T_~ ~ { r--~. "~ .,,..-,~<~::~,~-.~ ~, ..~
Experienced-Stimulated
Naive.Stimulated
Naive-Unstimulated
,..~?'=2- k=..~ ,,
((
/ •
/
fJ~-..
.~
,..-.,-.-.
x'~ i,, ~ )
.!ji,/
:)J!'
(.t.~_......._)_,_,~_.,~_.....__.
,f
..... ~
:)))) " ....
-,..
,'
~I
----.
/F,-"--r"-
-"'"c-"'-.~
",,,
It," i
\,,,~.,i/'---
-f'-''
,i
.~
..~
"
•,-- . . . . . ~ 1
.,_~.~ _ ~ , _ . ~ ' ,f--~'~,-~
• .'-..; ';..(',",. ,
I ("l ........... ,. . . . .
. .., .,_t.
_____. ..
i - . . . . . . . ~ I "-~'~-~'--
,--.~.,~-m J( ~ . . . \
I-'.L__./.,'~ '.'
..... ...--~___..--',-.: ..... ..,
LI.& i___ ,."
, .q,~.,.-
"\
...~
t,ti k.
\,,,t 1
ilL
.~S~/,,/'v~.~-~-,'c.-, ".
/'r~" $,/~ IL'~= ' h*"~;II': , t!
k
\,. .... .--Y-,. :.-/
/'lJ ~ ,,/
Fig. 3. Drawings of a mid-hippocampal section at level 3 from each of 5 animals in each group, showing the pixels in the hippocampal region that fell within an optical density window of the darkest (most active) 7-10% of pixels in the image. The stimulated side is left, except for the last section in the experienced group and the last two sections in the naive group.
differ (Table I). As shown in the bar graphs of Fig. 2, these results parallel those in the same structures at level 3. In addition, it is apparent from the unilateral means depicted in Fig. 2 that both CA1 and the ventral subiculum were comparably affected on the stimulated and unstimulated side. Again, an A N O V A performed on
the data from only the unstimulated side yielded the same pattern of significant results. There were no differences between groups in the dentate gyrus or the dorsal subiculum. This latter finding fails to support the suppression of the dorsal subiculum observed in both stimulated groups at level 3. T h e
204
Level 4 Naive-Unstimulated •
---
"
.1-?
---"i;
.. ..... T,-~f-~_:.-_~.. ,, ,, ,,/i "/. "'i: \\ '\
.................
~<;,.. ~'~..
i ,//,~,
,
,......,f--..--.J _~>. ~,
"~x\~
,, k
,:" ..~__~,,,~-.-..~. ,.
l/;:ir
,
i I(.~(
\"
-.
I'o~" '~
~ .
..... /
.,
~ x
t
Ig-~
I/
--~'~.- \
' /
r'
.,~,
~ i
",.~_ ~ _ ~
",____~.."
'",
~".
~
.,,",.~< /~/ r:~-//
........ 4b" ........ \..,YS~,-" ~-~.,\
~-~-~-~
-;
\
I
I
- ~_~
.... ~ . . . . . . .
ii
.1?~"'-*k _="
Y'/? I
....-.-"--~ .,.~:> .... IL ~"",, ,~- ,,'
-~...,,'.. '~,~ tl I
7, ...._.......( / +
Ii , i'~
i
._.__
Experienced.Stimulated
Naive-Stimulated
I
~J
--/
'\
Lb.:~'k
.,c.C_.~, /
¢
..
lj ''~ ,.,& \., -., \:~ \
\
/i
j, f
_
j,F~,
17
I ......
)' I,
~
-
-
",\..
i ././J ~
,3,...
°
~\',, ./.
J
Fig. 4. Drawings of a posterior hippocampal section at level 4 from each of 5 animals in each group, showing the pixels in the hippocampal region that fell within an optical density window of the darkest 7-10% of pixels in the image. The relative positions of brain sections correspond exactly to those in Fig. 3,
significant suppression in the retrosplenial cortex observed in the experienced animals at level 3 did not achieve significance in the corresponding measurements at level 4, although the means did show a trend to suppression (Table I). Foci o f activation To examine the anatomical localization of activation within the hippocampus and its variability within and
between groups, an optical density window routine was employed l°'n with the results shown in Figs. 3 and 4. After major landmarks were drawn, the window procedure produced a stippling that corresponds to the range of R O D values within the window. In this case the window was used to highlight the darkest 7-10% of pixels in the image (pixels with R O D values above the 90-93rd percentile), and then display those found in the hippocampai region. The larger proportion of darkened pixels
205 in the hippocampi of experienced-stimulated animals in Figs. 3 and 4 presents a clear anatomical demonstration of the significant quantitative effects reported above. As expected, the greatest activation appears in areas CA1 and CA3, with the dorsal-ventral extent of activation being broad and variable. Effects also appear to be more intense on the stimulated side (which is left, in the 4 experienced animals showing the largest effects). Also apparent is the overall absence of effect of stimulation in naive animals: in most cases, the naive-stimulated sections are indistinguishable from naive-unstimulated sections. Individual differences within groups may provide additional information related to the behavioral effects of stimulation. The last animal in the experienced-stimulated group shows considerably less activation compared to other animals in the same group at both levels 3 and 4, while the last animal in the naive-stimulated group shows relatively intense activation in CA1 at level 4 (see Figs. 3 and 4). Previous studies demonstrating that hippocampal stimulation pretreatment facilitates acquisition of hippocampal self-stimulation have frequently shown exceptional animals in the naive group that learned rapidly, unlike their cohorts 4'6. These exceptional animals may share common factors with the fifth animal shown in the naive-stimulated group in Fig. 4, such that stimulation results in wide propagation even in a naive animal. It has not been determined whether such factors might be extrinsic (such as electrode placement or vascular damage during implantation) or intrinsic (such as biochemical differences interacting with recruitment). In the two confirmed self-stimulators that were not stimulated during 2-DG uptake, measurements of areas CA1 and CA3 at levels 3 and 4 yielded mean ROD values that were within 1 S.E.M. of the means for the naive-unstimulated group given in Table I. This result is consistent with earlier studies in which rats previously kindled to behavioral convulsions showed no effect on 2-DG uptake in the absence of stimulation of the kindled amygdala 2 or the kindled hippocampus 15. Thus, the significant elevation of metabolic activity in the experienced-stimulated group can be attributed to the propagation of activity triggered by hippocampal stimulation administered during 2-DG uptake, and not to a long-term metabolic effect of previous stimulation experience alone. DISCUSSION The principal findings of this study indicate that hippocampal stimulation induces significantly more widespread propagation of activity in animals experienced with the stimulation through prior self-stimulation train-
ing than in naive animals. With the electrode loci and stimulation parameters employed in the present experiment, no evidence was found that over 500 stimulation trains delivered to the hippocampus in naive rats produced any significant metabolic activation of the structures examined at 1 mm anterior, or 2 or 3 mm posterior to the stimulating electrode. In contrast, experienced animals showed strong bilateral activation of the hippocampus proper and ventral subiculum at 2 and 3 mm posterior to the stimulating electrode. This result demonstrates that the 2-DG metabolic mapping technique is capable of revealing differential propagation due to use-induced plasticity even with modest stimulation parameters. The one previous report of increased 2-DG uptake in animals with prior hippocampal stimulation experience utilized a deliberately epileptogenic rapid kindling procedure 15. The present results provide experimental support for the neural correlate proposed in the hypothesis that hippocampal stimulation is not reinforcing in naive rats because of insufficient propagation, but acquires a reinforcing effect as stimulation experience engenders more widespread stimulation-induced activation. The results clearly indicate that even a large number of hippocampal stimulation trains in naive rats fail to propagate significantly within the hippocampus, while the same program of stimulation in experienced animals propagates widely throughout the hippocampus. While these findings support the contention of differential propagation as a function of experience, no direct inference can be made regarding the mechanism of this effect. The increased extent of metabolic labeling may be due to a potentiation-like mechanism, or to other neuroplastic effects of kindling. Studies demonstrating that stimulation pretreatment transfers its facilitative effect on acquisition of hippocampal self-stimulation to the contralateral hippocampus support trans-synaptic mechanisms over local excitability effects5. Possibly, multiple effects such as potentiation as well as decreased G A B A inhibition following hippocampal kindling 13 may contribute to the wider propagation of stimulationinduced activity observed in the hippocampus of experienced animals. Other effects that may contribute to the widespread propagation in experienced animals relate to the elicitation of epileptiform activity. Even though a low current level was chosen to minimize seizures, some seizure activity was elicited in experienced animals, but not observed in naive animals. Although AD was triggered by only the first stimulation train, it is possible that the widespread activation of the hippocampus observed in experienced animals is at least partly due to the initial epileptiform activity. Consequently, in the present ex-
206 periment, the naive and experienced groups differed not only in the presumed rewarding effect of the stimulation, but also in elicited A D activity. Thus, one cannot rule out the possibility that at least part of the increase in propagation observed in experienced animals is a correlate of hippocampal seizure rather than of hippocampal reward. Hippocampal reward need not be accompanied by epileptiform activity: previous studies have shown that maintenance of hippocampal self-stimulation does not depend on elicitation of A D activity 5. The pattern of activation resulting from hippocampal stimulation trains in experienced animals observed in the present experiment is in general agreement with previous studies of metabolic labeling during dorsal hippocampal stimulation. The most consistent finding is that unilateral hippocampal stimulation produces strong bilateral activation of hippocampal fields CA1 and CA314"15'23: This bilateral activation is consistent with the extensive commissural connections of the hippocampus, and with the electrode placement in CA3, from which the commissural fibers originate that project to contralateral CA1, CA3, and subiculum 22. Due to the more intense stimulation treatments administered in previous studies, extra-hippocampal activation reported in areas such as the amygdala 14"15 was not observed in the present experiment. However, the significant activation of the ventral subiculum at both levels 3 and 4 in the present experiment has been noted in only a few previous cases 14. There have been no previous reports of decreases in metabolic labeling resulting from stimulation of the hippocampus, but in the present experiment, an unex-
REFERENCES 1 Ackermann, R.E, Finch, D.M., Babb, T.L. and Engel Jr., J., Increased glucose metabolism during long-duration recurrent inhibition of hippocampal pyramidal cells, J. Neurosci., 4 (1984) 251-264. 2 Biackwood, D.H.R., Kapoor, V. and Martin, M.J., Regional changes in cerebral glucose utilization associated with amygdaloid kindling and electroshock in the rat, Brain Research, 224 (1981) 204-208. 3 Campbell, K.A., Development of the reinforcing effects of hippocampal stimulation through repeated daily electrical stimulation (kindling) (Doctoral dissertation, University of Toronto, 1982), Dissert. Abs. Internat., 43 (1983) 4189B. 4 Campbell, K.A. and Milgram, N.W., Potentiating hippocampal stimulation facilitates acquisition of lever-pressing for stimulation but not food, Physiol. Behav., 24 (1980) 1115-1118. 5 Campbell, K.A. and Milgram, N.W., Mechanisms underlying the plasticity of hippocampal stimulation-induced reward, Behay. Neurosci., 99 (1985) 209-219. 6 Campbell, K.A., Milgram, N.W. and Christoff, J.K., Plasticity in the reinforcing consequences of hippocampal stimulation, Brain Research, 159 (1978) 458-462. 7 Caudarella, M., Campbell, K.A. and Milgram, N.W., Hippocampal stimulation limits performance but does not retard acquisition of food-reinforced learning, Physiol. Behav., 32 (1984) 895-898.
pected consistent bilateral suppression of metabolic activity was observed in the posterior cingulate and retrosplenial cortex of experienced animals relative to the naive groups. This difference was observed at levels 1, 3 and 4, although the comparison was significant at levels 1 and 3 only. Because the R O D expresses metabolic activity of a structure relative to the image as a whole, it is conceivable that the suppression observed in the cingulate-retrosplenial cortex might be an artifact of the increase in activity over relatively large areas of the hippocampus in the experienced group. Observations mitigating against this interpretation are the fact that the suppression was significant at level 1 where no increases in metabolism were seen, and the lack of effect on other structures such as the dentate gyrus at levels 3 and 4, where large areas of activation were seen. Since CA3 is known to project to the posterior cingulate area 21, the suppression observed in the cingulate region in the experienced group may reflect inhibitory activity propagated as a result of stimulation. However, combined electrophysiologic and deoxyglucose labeling studies in the hippocampus have indicated that the metabolic demands of inhibitory interneurons can render paradoxical correlations between metabolic activity and excitation-inhibition of principal neurons 1. Acknowledgements. This work was supported by National Science Foundation Grant BNS 82 11972 to C.R. Gallistel; image processing was supported by National Institutes of Health Biotechnology Resource Center Grant 23-1352630. The author was supported on a postdoctoral fellowship from National Institute for Mental Health Training Grant T32 MH15092. The author would like to thank Elysa Braunstein for technical assistance.
8 Douglas, R.M. and Goddard, G.V., Long-term potentiation of the perforant path-granule cell synapse in the rat hippocampus, Brain Research, 86 (1975) 205-215. 9 Estes, W.K., The statistical approach to learning theory. In S. Koch (Ed.), Psychology: A Study of a Science, Vol. 2, McGrawHill, New York, 1959, pp. 380-491. 10 Gallistel, C.R., Gomita, Y., Yadin, E. and Campbell, K.A., Forebrain origins and terminations of the MFB metabolically activated by rewarding stimulation or by reward-blocking doses of pimozide, J. Neurosci., 5 (1985) 1246-1261. 11 Gallistel, C.R., Piner, C.T., Allen, T.O., Adler, N.T., Yadin, E. and Negin, M., Computer assisted analysis of 2-DG autoradiographs, Neurosci. Biobehav. Rev., 6 (1982) 409-420. 12 Gallistel, C.R. and Tretiak, O., Microcomputer systems for analyzing 2-deoxyglucose autoradiographs. In R.R. Mize (Ed.), The Microcomputer in Cell and Neurobiology Research, Elsevier, New York, 1985, pp. 390-408. 13 Kapur, J., Stringer, J.L. and Lothman, E.W., Evidence that repetitive seizures in the hippocampus cause a lasting reduction of GABAergic inhibition, J. Neurophysiol., 61 (1989) 417-426. 14 Kliot, M. and Poletti, C.E., Hippocampal afterdischarges: differential spread of activity shown by the [14C]deoxyglucose technique, Science, 204 (1979) 641-643. 15 Lothman, E.W., Hatlelid, J.M. and Zorumski, C.E, Functional mapping of limbic seizures originating in the hippocampus: a combined 2-deoxyglucose and electrophysiologic study, Brain Research, 360 (1985) 92-100.
207 16 Milgram, N.W., Server, A.C. and Campbell, K.A., Effect of food and water deprivation on hippocampal self-stimulation and post-stimulation feeding, Physiol. Psychol., 5 (1977) 43-48. 17 Olds, M.E. and Fobes, J.L., The central basis of motivation: intracranial self-stimulation studies, Ann. Rev. Psychol., 32 (1981) 523-574. 18 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic, New York, 1982, 162 pp. 19 Racine, R., Gartner, J.G. and Burnham, W.M., Epileptiform activity and neural plasticity in brain structures, Brain Research, 47 (1972) 262-268. 20 Racine, R., Newberry, E and Burnham, W.M., Post-activation potentiation and the kindling phenomenon, Electroencephalogr.
Clin. Neurophysiol., 39 (1975) 261-271. 21 Swanson, L.W. and Cowan, W.M., An autoradiographic study of the organization of the efferent connections of the hippocampal formation in the rat, J. Comp. Neurol., 172 (1977) 49-84. 22 Swanson, L.W., Sawchenko, P.E. and Cowan, W.M., Evidence for collateral projections by neurons in Ammon's horn, the dentate gyrus, and the subiculum: a multiple retrograde labeling study in the rat, J. Neurosci., 1 (1981) 548-559. 23 Watson Jr., R.E., Edinger, H.M. and Siegel, A., A p4C]2deoxyglucose analysis of the functional neural pathways of the limbic forebrain in the rat. II. The hippocampal formation, Brain Res. Rev., 5 (1983) 133-176.