Brain Research, 105 (1976) 389-403
389
© Elsevier ScientificPublishing Company, Amsterdam - Printed in The Netherlands
Research Reports
C O N N E C T I O N S OF T H E N U C L E U S A C C U M B E N S
E R V I N W. POWELL AND R O B E R T B. L E M A N
Department of Anatomy, University of Arkansas Medical Center, Little Rock, Ark. 72201 (U.S.A.) (Accepted September 2nd, 1975)
SUMMARY
Reports from previous works have given different classifications for the nucleus accumbens. There also appears to be a general lack of information regarding the fiber connections of the nucleus. The present investigation was undertaken to clarify the connections of this structure. Silver impregnation methods were used to discern some of the afferent fibers of the nucleus, and autoradiographic techniques were used to locate target areas of efferent projections. Afferents were found to be predominately from the septum. Other sources of possible afferents were the mid cingulate gyrus and the ventral nucleus of the diagonal band. No argyrophilia was observed in the nucleus accumbens following transection of the fornix body, lesions of the anterior orbital frontal cortex or anterior cingulate gyrus. On the basis of grain counts made from autoradiographic studies, the nucleus accumbens projects predominately to the lateral hypothalamus. Counts above background were found in the cingulate gyrus, septum, ventral nucleus of the diagonal band, midline thalamic nuclei, habenula, caudate and substantia nigra. Thus, efferent projections appear to distribute to both limbic and extrapyramidal structures. Considering these connections and the functions reported by various workers the nucleus accumbens may serve as a bridge between limbic and extrapyramidal motor systems effecting limbic influence in some movements.
INTRODUCTION
The nucleus accumbens (NA) has been suggested by various workers as a part of functionally different systems. This is understandable, since the nucleus ontogenetically and phylogenetically differentiates from the area between the septum and the striatum 43. Some basis for this diversity depends upon the connections delineated by
390 each investigator. Most have classified the NA as part of the striatum, i.e., the medial extension of the head of the caudate nucleusS,11,18, 22. Others have reported it to be part of the lateral paraolfactory areaS,1L Stephan and Andy 42 did not list the NA in their extensive definition of the septum; however, Crosby et al. ~, Kappers and Theunissen ~8 and Lauer 22 reported that the NA should be classified in such a manner. Functionally, the nucleus has been linked with various roles such as visceral responsesaa, 24, olfaction 17, milk ejection 44, alerting the limbic system 14 and somatomotor activity 29. There also appears to be a general lack of information regarding fiber connections of the nucleus. The purpose of this study was to assess what is known about connections of the NA and to investigate some of these connections in the squirrel monkey. Autoradiographic mapping techniques were used to determine efferent targets, and silver impregnation methods were used to determine afferent origins. METHODS
Twenty-nine monkeys were used in this study. Each monkey was anesthetized by an intramuscular injection of ketamine hydrochloride (25 mg/kg) which was augmented every 20-30 min. The animals were oriented in a stereotaxic instrument for the accurate positioning of lesions or for the injections of radioactive leucine. A local anesthetic was injected into the scalp preliminary to surgery (Xylocaine, 2 ~, containing epinephrine).
(A) Efferents Target nuclei of the efferent connections of the NA were examined by autoradiographic methods. A 10-/A Hamilton syringe with a 26-gauge needle mounted on an electrode carrier was lowered to the coordinates of the NA (A13, L1, H1)s. One #1 of L-[4,5-ZH]leucine dissolved in normal saline (New England Nuclear Corp., concentrated to 10/~Ci//~l; spec. act., 5 Ci/mmole) was administered over a 5-min period (0.2 #l/min). After infusion of the total injection (l #1), the syringe was left in place for 5 additional min before withdrawal. The syringe plunger was advanced by a micrometer drive which allowed fine control of delivery. The patency of the needle opening and operation of the injection assembly was tested and washed clean before injections were made. The slow, careful injection restricts tissue damage which might be caused by internal fluid pressure. Leaving the syringe in place for an additional 5 min allows time for the labeled leucine to be taken up by the neuronal tissue and thus reduced the chances of spread along the needle tract. Following a survival time of 1 day, the monkeys were deeply anesthetized with sodium pentobarbital, and the brain was perfused through the left ventricle of the heart with 10 ~ formalin in 0.9 ~ saline. Subsequently, the brains were removed, stored in 10 ~ formalin, and blocked in three parts for paraffin embedding. Sections (8 #m in thickness) were selected every 150 #m from the genu to the splenium of the corpus callosum. These were mounted on slides which subsequently were dipped back to back in undiluted Kodak NTB-2 emulsion. Following exposure periods of 2, 4, 6 and 12 weeks, the slides were developed and fixed. The slides were stained with cresyl violet using a method modified from those reported by
391 Hendrickson et al. 15. Labeling of the perikarya and neuropil was compared at a magnification of × 1000. Random grain counts were made using an ocular grid of 400 squares, with a total area of 0.01 sq.mm. This grid was also used to measure the dimension of the injection site.
(B) Afferents Sources of afferents of the NA were examined using the Fink-Heimer I and Nauta-Gygax methods of labeling degenerating axonsg,zT. Cases were selected to examine degeneration within the NA following lesions in the septum (10 cases), cingulate gyrus (8 cases), hippocampus (3 cases), orbital frontal cortex (2 cases), ventral nucleus of the diagonal band (1 case), and caudate nucleus (1 case). The nomenclature of the septal nuclei has been adopted from Stephan and Andy4L For silver impregnation studies, either electrolytic lesions or fornix transections (scissor cut using offset spear scissors mounted on an electrode carrier) were placed stereotaxicallyzl. Following a 7-day postoperative period, the animals were perfused as previously stated. Later, frozen coronal sections (25 #m) were cut and stained. RESULTS
The nucleus accumbens (NA) is a large group of cells extending from the bed nucleus of the stria terminalis to levels anterior to the genu of the corpus caUosum and the anterior olfactory nucleus. The nucleus lies ventral to the anterior horn of the lateral ventricle between the diagonal band and the arm of the anterior limb of the anterior commissure. There appear to be two components to the nucleus. The lateral part lies between the ventral tip of the anterior horn of the lateral ventricle and the putamen. The cells are interspersed with transversing fibers and are somewhat scattered. The medial more cellular part extends from the ventral tip of the anterior horn of the lateral ventricle to the island of Calleja and the diagonal band. This medial part is the primary area of this study.
(A) Efferents In two of the animals the injection of labeled leucine was reasonably confined to the limits of the NA. The data indicating terminals in target nuclei for these animals were nearly identical. One of the injection sites is illustrated in Fig. 1. During the injection of radioactive leucine, some of the amino acid would likely be conveyed outward from the center of the injection site by simple diffusion. Previous studies have shown that labeled amino acid is preferentially absorbed by cell bodies and is not taken up by axons to any great extent 7. Thus, the leucine injected for this study was absorbed by perikarya around the injection center. In the prepared slides, labeled leucine is made to appear as silver grains representing amino acid bound as protein. Moreover, the absorption time of the labeled leucine is sufficiently short to indicate that most or all of the injectate was absorbed by perikarya at the injection site and was not later conveyed to other areas of the brain. Within the injection site the majority of perikarya are covered with silver grains
392
A Fig. 1. Illustrations of injection site for radioactive leucine. The stippling in A indicates the location of the injection in the nucleus accumbens for S-76. The center photograph indicates the area of symmetrical diffusion at the injection site. C, taken from the inset in B, shows an intense grain density over neuronal somata as well as between neurons. For abbreviations used in this and followingfigures see p. 401.
(Fig. 1C). Outside each of the blackened cell bodies, whorled tracts of grains are visible. These tracts may overlie proximal axons, and possibly dendrites, packed with radioactive proteins. As one moves from the center of the injection site toward the periphery, the number of blackened cells decreases as the concentration of grains overlying cell bodies and neuropil diminishes. The amount of labeling over perikarya decreases until the concentration of grains over them equals that observed over the surrounding neuropil. We have considered this to be the end of the diffusion area at the site of injection. We have defined 'diffusion space' as that area in and around the injection site in which perikaryal labeling equals or exceeds neuropil labeling. The diffusion space illustrated in Fig. 1 extended from the ventral tip of the right lateral ventricle 3.35 mm to the ventral border of the brain, possibly including a portion of the olfactory tubercle; from the isle of Calleja, it extended laterally 1.48 mm; in an antero-posterior dimension, the diffusion space extended 1.60 mm (A-P 15.3 to A-P 13.7). This shape was largely due to the slant of the needle tip which made the opening ovoid. The needle tract also may have permitted some longitudinal spread along its penetration. The volume of this ellipsoida| diffusion space was calculated to be 7.54 #1. The total number of cells within the diffusion space of the nucleus accumbens, calculated to be available to absorb the radioactive leucine, was 190,000 cells. Terminal projections of the NA, as revealed by the autoradiographic procedures, are illustrated for several areas in Figs. 2 and 3. The greatest density of grains was found in the lateral preoptic area. The grain pattern in the lateral external nucleus of the septum and lateral hypothalamus had the next highest density (Fig. 3). Grains
393
/
W-, /
j
f
.-,!~..... •.
~
i/////J~
~~.,~7 ',I 0 ' ~
~4 ~
~,
.....
/
F@ ,=L Fig. 2. Pattern of silver grains representative of terminal endings, following injection of radioactive leucine into the nucleus accumbens (A). The intense stippling in the nucleus accumbens illustrates the injection site for S-75. This section is 0.5 mm posterior to the center of the injection. All other stippling (grain patterns) in the illustrations from A to F represent terminals. B is 2 mm posterior to A.
were f o u n d t h r o u g h o u t the septal nuclei. L a b e l i n g was o b s e r v e d in the a n t e r i o r h y p o t h a l a m u s including the p a r a v e n t r i c u l a r a n d s u p r a c h i a s m a t i c nuclei. The lateral a n d d o r s a l h y p o t h a l a m u s c o n t a i n e d labeling as far p o s t e r i o r as the m a m m i l l a r y b o d y . G r a i n s were o b s e r v e d in the s u b s t a n t i a i n n o m i n a t a as far l a t e r a l as the e n t o p e d u n c u l a r nucleus b u t were n o t o b s e r v e d to be a b o v e b a c k g r o u n d levels m o r e l a t e r a d or in the a m y g d a l a . A restricted p a r t o f a r e a 24 o f the cingulate gyrus a n d the a n t e r i o r tip o f the h i p p o c a m p u s n e a r the uncus also c o n t a i n e d a n u m b e r o f grains. G r a i n s i n d i c a t i n g sites o f a x o n terminals were f o u n d to be o f a m o r e restrictive p a t t e r n in the t h a l a m u s ,
394
Fig. 3. Pictures of target structures demonstrating the density of silver grains present after a 12-week exposure period. Background activity was not significantly increased during this exposure time as compared with sections exposed for 6 weeks. However, the target areas were more easily distinguished at 12 weeks of exposure. Hippocampal gyrus (Hg) represents a control, i.e., background level of grain density at 12 weeks.
395
Fig. 4. Additional controls, i.e., examples of the density of background activity as represented by grain concentration at 12 weeks of exposure (S-76). NA, contralateral nucleus accumbens; Cg, ipsilateral cingulate gyrus area 23; Pu, ipsilateral putamen; Fa, ipsilateral fastigial nucleus. Thalamic nuclei over which grain counts exceeded that of control areas were the parataenial, periventricular, reuniens and dorsal medial. The habenular nuclei were the only other diencephalic structures to show labeling. The caudate nucleus (the medial part of the head and anterior part of the body) and pre-rubral field were the heaviest labeled extrapyramidal areas. Grains observed over the pars compacta of the substantia nigra were slightly less than those found in the thalamus. The globus pallidus had little or no labeling in the sections studied.
396
Fig. 5. Composite illustration of silver impregnation data showing representative lesions (top row across); subsequent argyrophilic pattern in the N A for each type of lesion (center row), and photomicrographs of the silver impregnated fibers for the cases used (bottom row). The pictures in the two left frames were taken from sections using the Fink-Heimer I technique for cases S-63 and S-38. The picture in the bottom right corner was taken from a section stained using Nauta-Gygax technique, case S-22. The calibration line applies only to the 3 photomicrograhs. 0.7 centimeter equals approximately 1 mm in the diagrams.
397 Grains were not above background activity in the amygdala, contralateral NA, fastigial nucleus and thalamic nuclei other than those reported above. Examples of control areas and background counts are shown in Fig. 4.
(B) Afferents The argyrophilia observed in the NA following lesions in the septum, cingulate gyrus and diagonal band differed in amount and in pattern of distribution. Following electrolytic lesions of the septum, argyrophilia was consistently observed within the NA (Fig. 5). The lesions were located dorsal to the anterior commissure and within the near margin of the corpus callosum from A13 to A 15. The electrode tract penetrated the overlying cingulate gyrus and superior frontal gyrus. Some component of the dorsal septal nuclei was involved in each case, i.e., the anterior, external, internal or intermediate parts. The degeneration patterns in the NA for these cases did not differ topographically from one another. The degeneration pattern was strongest in the medial and ventral portions of NA. Lesions of the cingulate gyrus were located from A8 to A10 in Brodmann's area 24. The lesion illustrated was small and largely confined to the cingulate gyrus. Other cases sustained larger lesions involving damage to the overlying frontal gyrus. However, greater amounts of argyrophilia were not observed within the NA for these latter cases. Therefore, the more anatomically confined lesion is shown in Fig. 5. Degeneration was present in the central area and on the medial edge of the nucleus. The overall density of argyrophilia was slightly less than it was following the septal lesion. The lesion in one animal was located in the diagonal band near the olfactory tubercle area (A14) but did not involve the medial part of the NA (Fig. 5). Heavy impregnation of fibers was seen on the medial edge of the NA (Fig. 5). The argyrophilic pattern was restricted to a narrow band along the medial edge and ventral border of the NA. Degeneration was found in the NA following a lesion of the medial part of the head of the caudate nucleus. This lesion had septal involvement, and the density or pattern of degeneration in the nucleus accumbens was not significantly different from that observed following lesions confined to the septum. The NA contained no argyrophilic particles following transection of the fornix, small lesions of the anterior cingulate gyrus, or lesions of the orbital-frontal cortex. The fornix of three animals was unilaterally sectioned (A7, L2.5, H + 9) using a stereotaxically placed extended scissors. The transection appeared to be complete in one case. The second animal's transection was similar but the degeneration in the lateral tip was less intense. In the third case, the lateral tip was free of degeneration. A large amount of argyrophilia was observed in the fornix column, anterior nuclei of the thalamus and mammillary body. No evidence of degeneration was seen in the NA in sections of any of the 3 animals although degeneration was obvious in main target structures in the thalamus (anterior nuclei, dorsal medial nucleus and dorsal lateral nucleus) and hypothalamus, especially the mammillary nuclei. Lesions of the anterior cortex were made in 4 animals by suction. All involved
398 the frontal gyrus with encroachment of the dorsal one-fourth of the cingulate gyrus, Brodmann's area 24 (A!8-19). A lesion was made by suction in the anterior half of the orbital-frontal cortex of two animals. Equal amounts of the overlying frontal lobe were also removed to gain access. Fiber and preterminal labeling were obvious in the internal capsule and dorsomedial nucleus of the thalamus. However, none was observed in the NA from either case. DISCUSSION
The results show that the NA is connected to the lateral hypothalamus, other limbic structures and extrapyramidal structures. The location of the NA would allow it to be related morphologically to either the caudate or septal nuclei. However, most of the results demonstrate that the connections of the NA are predominately within the hypothalamus and limbic system.
(A) Efferents Various investigators have described both slow and fast components of axoplasmic transport of incorporated radioactive leucine 7,za. Data of the efferent connections of this study are based upon the fast component. If one assumes that all of the free amino acid is taken up before the injection needle is removed, little if any labeled material would follow the tract during and after its withdrawal. The total amount of labeled leucine injected was 0.2 nmoles, administered in 0.04 nmole doses every minute for 5 rain. Five additional minutes were allowed before withdrawal of the needle. Assuming that all of the leucine transport proteins were maximally operative during the entire injection, the rate of transport of this amino acid across the neuron membrane equalled 0.9 nmoles transported per minute or the absorption time would be 15 sec. If non-saturated conditions are assumed, approximately 11 min would be required for absorption of the total injection of labeled leucine. The pulsed nature of the injection favors maximum velocity of uptake; therefore, the time span allotted is sufficiently long to expect that all of the labeled leucine was absorbed prior to removal of the needle. This would minimize the possibility of labeled tracer creeping along the needle tract and inadvertently labeling additional structures. This rate of uptake also strengthens the point that grain counts above background levels in nuclei beyond the bounds of the diffusion space (7.54/A) are terminal sites which have accumulated the labeled tracer. However, we were not able to differentiate between possible diffusion of isotope and what we believed to be short axon projections. Neuronal injury at the site of injection may have occurred as a result of introducing the 26-gauge needle into the N A tissue. The number of cells injured compared to the total number of cells in the diffusion space was so small that even if all the cells over the width of the needle tract died (approx. 8000) it would not decrease the total volume sufficiently to negate the validity of the above calculations. Our data show that the lateral hypothalamus is one of the major target sites of
399 the NA. Various investigators 11,19,a7 have reported this connection and described the pathway as coursing through the medial forebrain bundle. The results indicate that projections of the NA are topographic to the anterior cingulate gyrus. Powel130 traced degeneration to this portion of the cingulate gyrus in the rat after placing a lesion in the lateral septal nuclei which did overlapping damage to the NA. The results indicate a general projection to the septal nuclei. This is in partial agreement with studies using lesions which had slight encroachment on the NA19,22, 44. These workers reported degeneration within a limited septal region. Our injections, while directed to the medial half of the NA, were largely confined to the NA and occupied a large part of the nucleus. Connections from the NA were found to be distributed to the diagonal band within the olfactory tubercle and to structures within the diencephalon, i.e., thalamic midline nuclei (parataenial, periventricular, reuniens and dorsal medial) and the medial habenular nuclei. Authorsla, 19 have reported a substantial connection from the NA, to the medial part of the dorsal medial nucleus. We believe that the extra amount of degeneration present in their work was probably due to surgical involvement of the septum or more likely to fiber paths from the orbital-frontal cortex 31. The relatively high counts in the medial habenular nucleus appear to arise in the NA since the olfactory tubercle region projects preferentially to the lateral nucleus 28. Our data on non-limbic targets of the NA support the work of others who have also reported target nuclei to be the caudate ~2 and substantia nigra a7,41. However, we were unable to confirm connections with the globus pallidus as reported by Smith 41. (B) Afferents
Most of the afferent connections of the medial compact portion of the NA come from the septum (a limbic structure). This is in agreement with data reported by several authorsa0,z6, 3s. The fibers appear to originate in the septum and not from fornix fibers traversing this region, since no degeneration was found in the NA following transection of the fornix. Some authorsl,3,10,19 have reported degeneration in the NA after fornix section, but extensive surgical damage was shown in their studies. In a recent study using Fink-Heimer I silver impregnation methods, in the monkey, Seigel et al. ~9 observed terminations in the NA following hippocampal lesions using a 2-day postoperative survival period. It is possible that the terminals are structured in such a way (size?) that they are not visible after a 7-day postoperative period as used in our study. Also, in spite of the intensity of degeneration in the fornix column, thalamus and hypothalamus, it is possible that the lateral tip of the fornix was spared in all 3 animals in our transections of the fornix. Seigel et al. 39 reported that the hippocampus projected to the NA via the lateral fibers of the fornix. The cingulate gyrus (limbic cortex) had the next largest afferent connection to the medial part of the NA. This connection demonstrated a definite topography since only those cingulate lesions within a restrictive part of area 24 produced degeneration in the NA. The rather limited more anterior lesions of the cingulate gyrus were
400 found to give negative results. Krieg 20 states that area 24 of the cingulate gyrus probably has a variety of specific topography in the majority of its subcortical projections. Such highly topographic relations may be the reason why Domesick 6 and Powell et al. 34 did not report this connection. The diagonal band part of the olfactory tubercle, another timbic region, was shown to project to the NA. A few authors 2~,30,44 have reported this connection to be more extensive than our results indicate. The quantitative representation of this connection in our study may appear to be deficient because of the staining limitations of the N a u t a - G y g a x method, i.e., fine terminals do not stain well 32. The results indicate that the anterior half of the orbital frontal cortex does not project to the NA. More posterior areas of the gyrus were avoided because of the juxtaposition of the NA. Studies using silver impregnation methods 19 and electrophysiological data (evoked potentials, 4 msec latency) indicate that the amygdala may be monosynaptically related to the nucleus accumbens 33. This is in agreement with data (antidromic responses) collected from amygdaloid units following stimulation of the nucleus accumbens 16. Other origins of the N A afferents have been reported by various workecs. Projeztions to the N A were found to arise from thalamic structures, e.g., dorsal medial nucl~us (limbic) 23, parataenial nucleus2,35, 40 and parafascicular nucleus 4°. Guillery 12 reported finding projections to the NA from the hypothalamus. Connections of the N A are not limited to limbic structures; however, these few connections (non-limbic) occurred with extrapyramidal structures. Our work on the NA afferents had only one lesion in a non-limbic structure, i.e., the caudate nucleus. This lesion had involvement of the septum, and therefore the origin of the degeneration could not be determined precisely. K n o o k 19 also had the problem of septal involvement and was inconclusive about caudatoaccumbens connections. Other afferents were reported from substantia nigra 2~ and fastigial nucleus of the cerebellum 2s. CONCLUSION
The combined afferent and efferent data are summarized in Fig. 6. Our data indicate that most interconnections are between limbic structures and the NA. There is a cleat and extensive connection with the lateral hypothalamus and the septum. O f the non-limbic structures discussed, all except the cerebellum are considered to be within the extrapyramidal system. The extrapyramidal system and cerebellum are both functionally related to somatic muscle control. Data from recent experimentation suggest to us that the N A is involved in motor function 29 and may even differentiate in conjunction with motor demands 4. Considering that the connections are with limbic and extrapyramidal structures, the NA may serve as a bridge between limbic and extrapyramidal m o t o r systems effecting limbic influence in motor behavior. Eating and chewing, for example, are mixed visceral and m o t o r activities typical of limbic system involvement in phenomena like hunger, thirst, and fight behaviors. The
401
Fig. 6. S u m m a r y of connections of the N A . Solid lines represent efferent connections based on the autoradiographic data of this study. The dotted lines represent the afferent connections based on the data obtained f r o m the silver impregnation studies. The dashed lines represent afferent connections described by other authors.
muscles controlling jaw movement are phylogenetically and ontogenetically derived from the pharyngeal arches. Furthermore, temporal lobe seizure activity commonly involves jaw movement and salivation which are primarily visceromotor manifestations. LIST OF ABBREVIATIONS USED Ac = AmAv = Cc = Cd = Cg = Cp = Db = Dh = Dm = Ds = Fa = Fx = Gp = Hg = HI = Hm =
anterior commissure amygdala anteroventral nucleus of the thalamus corpus callosum caudate cingulate gyrus cerebral peduncle diagonal band dorsal h y p o t h a l a m u s dorsal medial nucleus o f the thalamus dorsal septal nucleus fastigial nucleus fornix globus pallidus hippocampal gyrus lateral habenula medial habenula
Hp Ic Ig Ld Lh Ls Mb Mh Ms NA Oc Of Op Ot Pa Pr
= = = = = = = = = = = = = = =
hippocampus internal capsule indusium griseum lateral dorsal nucleus of the thalamus lateral h y p o t h a l a m u s lateral septum mammillary body medial h y p o t h a l a m u s medial septal nucleus nucleus accumbens optic chiasm orbital frontal cortex optic tract olfactory tubercle paraventricular nucleus of the hypothalamus = pre-rubral field
402 Pt parataenial nucleus of the thalamus Po preoptic area Pu - putamen Pv - periventricular nucleus of the thalamus Sc suprachiasmatic nucleus of the hypothalamus Se :, septum Sn substantia nigra Sm stria medullaris thalami
Ti Th Un V Va
temporal lobe thalamus uncus ~ ventricle ventral anterior nucleus of the thaiamus Vh .... ventral hypothalamus Zi zona incerta
ACKNOWLEDGEMENTS T h i s s t u d y was s u p p o r t e d by N S F G r a n t G B - 3 2 1 7 0 . T h e a u t h o r s wish to t h a n k D r . S. A. G i l m o r e f o r h e r h e l p w i t h t h e l a b o r a t o r y p r o c e d u r e s f o r a u t o r a d i o g r a p h y , Dr. D o n D e L u c a f o r his assistance w i t h c o m p u t a t i o n s a n d P a u l R o b i n s o n for his t e c h n i c a l aid.
REFERENCES 1 CARMAN,J. B., COWAN, W. M., AND POWELL, T. P. S., The organization of cortico-striate connexions in the rabbit, Brain, 86 (1963) 525-562. 2 COWAN, W. M., AND POWELL, T. P. S., The projection of midline and intralaminar nuclei of the thalamus of the rabbit, J. NeuroL Neurosurg. Psychiat., 18 (1955) 266-279. 3 CRAGG, B. G., AND HAMLYN, L. H., Histological connections and visceral actions of components of the fimbria in the rabbit, Exp. Neurol., 2 (1960) 581-597. 4 CREPES, E. S., Time of origins in preoptic and septal areas of the mouse: an autoradiographic study, J. comp. NeuroL, 157 (1974) 161-244. 5 CROSBY, E. G., HUMPHREY, T., AND LAUER, E. W., Correlative Anatomy of the Nervous System, Macmillan, New York, 1962, pp. 360 and 419. 6 DOMESICK, V. B., Projections from the cingulate cortex in the rat, Brain Research, 12 (1969) 296-320. 7 EDWARDS,S. B., The ascending and descending projections of the red nucleus in the cat: an experimental study using an autoradiographic tracing method, Brain Research, 48 (1972) 45-63. 8 EMMERS, R., AND AKERT, K., A Stereotaxic Atlas of the Brain of the Squirrel Monkey (Sahnini sciureus), University of Wisconsin Press, Madison, Wisc., 1963, pp. 3 ft. 9 FINK, R. P., AND HEIMER, L., Two methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system, Brain Research, 4 (1967) 369-374. 10 Fox, C. A., The stria terminalis, longitudinal association bundle and precommissural fornix fibers in the cat, J. comp. NeuroL, 79 (1943) 277-296. 11 GURDJIAN,E. S., The corpus striatum of the rat, J. comp. NeuraL, 45 (1928) 249-281. 12 GUILLERY,R. W., Degeneration in the hypothalamicconnexion of the albino rat, J. Anat. (Land.), 91 (1957) 91-115. 13 GUILLERY,R. W., Afferent fibers to the dorso-medial thalamic nucleus in the cat, J. Anat. (Lond.), 93 (1959) 403~,19. 14 HEATH, R. G., Physiologic basis of emotional expression: evoked potential and mirror focus studies in rhesus monkeys, Biol. Psychiat., 5 (1972) 15-31. 15 HENDRICKSON,A., MOE, L., AND NOBLE, B., Staining for autoradiography of the central nervous system, Stain Technol., 47 (1972) 283-290. 16 ITO, N., ISHIDA, H., M1GAKAWA,F., AND NAITO, H., Microelectrode study of projections from the amygdaloid complex to the nucleus accumbens in the cat, Brain Research, 67 (1974) 338-341. 17 JOHNSTON,J. B., Further contributions to the study of the evolution of the forebrain, J. comp. NeuroL, 35 (1923) 337-382. 18 KAPPERS, C. U., UND THEUNISSEN,W. F., Die Phylogenese des Rhinencephalons, des Corpus striatum, und der Vorderhirnkommissuren, Folia neuro-biol. (Lpz.), 1 (1970) 173-288.
403 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
KNOOK, H. L., The Fibre Connections of the Forebrain, Davis, Philadelphia, Pa., 1966, 145 pp. KREIG, W. J. S., Connections of the Cerebral Cortex, Brain Books, Evanston, II1., 1963, 219 pp. LAS~K, R. J., Protein transport in neurons, Int. Rev. Neurobiol., 13 (1970) 289-324. LAUER, E. W., The nuclear pattern and fiber connections of certain basal telencephalic centers in the macaque, J. comp. Neurol., 82 (1945) 215-254. LEONARD, C. M., The prefrontal cortex in the rat. (I) Cortical projection of the mediodorsal nucleus. (II) Efferent connections, Brain Research, 12 (1969) 321-343. LORENS,S. A., SORENSEN,J. P., ANO HARVEY,J. A., Lesions in the nuclei accumbens septi of the rat: behavioral and neurochemical effects, 3. comp. physioL PsychoL, 73 (1970) 284-290. MOORE,R. Y., BHATNAOAR,R. K., AND HEELER, A., Anatomical and chemical studies of a nigroneostriatal projection in the cat, Brain Research, 30 (1971) 119-135. NAUTA, W. J. H., Hippocampal projections and related neural pathways to the mid-brain in the cat, Brain, 81 (1958) 319-340. NAUTA, W. J. H., AND GYGAX, P. A., Silver impregnation of degenerating axons in the central nervous system. A modified technique, Stain Technol., 92 (1954) 91-93. PAUL, S. M., HEATH, R. G., AND ELLISON,J. P., Histochemical demonstration of a direct pathway from the fastigial nucleus to the septal region, Exp. Neurol., 40 (1973) 798-805. PLINENBURG,A. J. J., WOODRUFF,G. N., AND VAN ROgSUM,J. M., Ergometrine induced locomotor activity following intracerebral injection into NAS, Brain Research, 59 (1973) 289-302. POWELL,E. W., Septal efferents revealed by axonal degeneration in the rat, Exp. Neurol., 8 (1973) 406-422. POWELL, E. W., Limbic projections to the thalamus, Exp. Brain Res., 17 (1973) 394-401. POWELL, E. W., AND SCHNURR,R., Silver impregnation of degenerating axons; comparisons of postoperative intervals, fixatives and staining methods, Stain TechnoL, 47 (1972) 95-100. POWELL, E. W., CLARK, W. M., AND MAUKAWA,J., An evoked potential study of limbic projections to nuclei of the cat septum, Electroenceph. clin. Neurophysiol., 25 (1968) 266-273. POWELL,E. W., AKAGI, K., AND HATTON,J. B., Subcortical projection of the cingulate gyrus in the cat, J. Hirnforsch., 15 (1974) 313-322. POWELL, T. P. S., AND COWAN, W. M., A study of thalamostriate relations in the monkey, Brain, 79 (1956) 364-390. RAISMAN,G., The connexions of the septum, Brain, 89 (1966) 317-368. RIOCH, D. M., Certain myelinated fiber connections of the diencephalon of the dog, cat and aevisa, J. comp. NeuroL, 53 (1931) 319-388. SEIGEL, A., AND TASSONI, J. P., Differential efferent projections of the lateral and medial septal nuclei to the hippocampus in the cat, Brain Behav. Evolut., 4 (1971) 201-219. SEIGEL, A., OHGAMI, S., AND EDINGER, O. i . , Projections of the hippocampus in the squirrel monkey, Presented at Society of Neuroscience 4th Annual Meeting, (1974). SIMMA,K., g u t Projektion des Centrum medianum und Nucleus parafascicularis thalami beim Menschen, Mschr. Psychiat. NeuroL, 122 (1951) 32--46. SMITH,O. C., The corpus striatum, amygdala, and stria terminalis of Tamandua tetradactyla, J. comp. NeuroL, 51 (1930) 65-127. STEPHAN, H., AND ANDY, O. J., Cytoarchitectonics of the septai nuclei in old world monkeys (Cercopithecus and Colobus), J. Hirnforsch., 7 (1964) 1-23. TUCHMANN-DUPLESSIS,H., AUROUX,i . , AND HAEGEL,P., Illustrated Human Embryology, Vol. 3, Nervous System and Endocrine Glands, Springer, New York, 1974, pp. 60-71. WOODS,W. H., HOLLAND,R. C., AND POWELL, E. W., Connections of cerebral structures functioning in neurohypophysial hormone release, Brain Research, 12 (1969) 26-46.
389
© Elsevier ScientificPublishing Company, Amsterdam - Printed in The Netherlands
Research Reports
C O N N E C T I O N S OF T H E N U C L E U S A C C U M B E N S
E R V I N W. POWELL AND R O B E R T B. L E M A N
Department of Anatomy, University of Arkansas Medical Center, Little Rock, Ark. 72201 (U.S.A.) (Accepted September 2nd, 1975)
SUMMARY
Reports from previous works have given different classifications for the nucleus accumbens. There also appears to be a general lack of information regarding the fiber connections of the nucleus. The present investigation was undertaken to clarify the connections of this structure. Silver impregnation methods were used to discern some of the afferent fibers of the nucleus, and autoradiographic techniques were used to locate target areas of efferent projections. Afferents were found to be predominately from the septum. Other sources of possible afferents were the mid cingulate gyrus and the ventral nucleus of the diagonal band. No argyrophilia was observed in the nucleus accumbens following transection of the fornix body, lesions of the anterior orbital frontal cortex or anterior cingulate gyrus. On the basis of grain counts made from autoradiographic studies, the nucleus accumbens projects predominately to the lateral hypothalamus. Counts above background were found in the cingulate gyrus, septum, ventral nucleus of the diagonal band, midline thalamic nuclei, habenula, caudate and substantia nigra. Thus, efferent projections appear to distribute to both limbic and extrapyramidal structures. Considering these connections and the functions reported by various workers the nucleus accumbens may serve as a bridge between limbic and extrapyramidal motor systems effecting limbic influence in some movements.
INTRODUCTION
The nucleus accumbens (NA) has been suggested by various workers as a part of functionally different systems. This is understandable, since the nucleus ontogenetically and phylogenetically differentiates from the area between the septum and the striatum 43. Some basis for this diversity depends upon the connections delineated by
390 each investigator. Most have classified the NA as part of the striatum, i.e., the medial extension of the head of the caudate nucleusS,11,18, 22. Others have reported it to be part of the lateral paraolfactory areaS,1L Stephan and Andy 42 did not list the NA in their extensive definition of the septum; however, Crosby et al. ~, Kappers and Theunissen ~8 and Lauer 22 reported that the NA should be classified in such a manner. Functionally, the nucleus has been linked with various roles such as visceral responsesaa, 24, olfaction 17, milk ejection 44, alerting the limbic system 14 and somatomotor activity 29. There also appears to be a general lack of information regarding fiber connections of the nucleus. The purpose of this study was to assess what is known about connections of the NA and to investigate some of these connections in the squirrel monkey. Autoradiographic mapping techniques were used to determine efferent targets, and silver impregnation methods were used to determine afferent origins. METHODS
Twenty-nine monkeys were used in this study. Each monkey was anesthetized by an intramuscular injection of ketamine hydrochloride (25 mg/kg) which was augmented every 20-30 min. The animals were oriented in a stereotaxic instrument for the accurate positioning of lesions or for the injections of radioactive leucine. A local anesthetic was injected into the scalp preliminary to surgery (Xylocaine, 2 ~, containing epinephrine).
(A) Efferents Target nuclei of the efferent connections of the NA were examined by autoradiographic methods. A 10-/A Hamilton syringe with a 26-gauge needle mounted on an electrode carrier was lowered to the coordinates of the NA (A13, L1, H1)s. One #1 of L-[4,5-ZH]leucine dissolved in normal saline (New England Nuclear Corp., concentrated to 10/~Ci//~l; spec. act., 5 Ci/mmole) was administered over a 5-min period (0.2 #l/min). After infusion of the total injection (l #1), the syringe was left in place for 5 additional min before withdrawal. The syringe plunger was advanced by a micrometer drive which allowed fine control of delivery. The patency of the needle opening and operation of the injection assembly was tested and washed clean before injections were made. The slow, careful injection restricts tissue damage which might be caused by internal fluid pressure. Leaving the syringe in place for an additional 5 min allows time for the labeled leucine to be taken up by the neuronal tissue and thus reduced the chances of spread along the needle tract. Following a survival time of 1 day, the monkeys were deeply anesthetized with sodium pentobarbital, and the brain was perfused through the left ventricle of the heart with 10 ~ formalin in 0.9 ~ saline. Subsequently, the brains were removed, stored in 10 ~ formalin, and blocked in three parts for paraffin embedding. Sections (8 #m in thickness) were selected every 150 #m from the genu to the splenium of the corpus callosum. These were mounted on slides which subsequently were dipped back to back in undiluted Kodak NTB-2 emulsion. Following exposure periods of 2, 4, 6 and 12 weeks, the slides were developed and fixed. The slides were stained with cresyl violet using a method modified from those reported by
391 Hendrickson et al. 15. Labeling of the perikarya and neuropil was compared at a magnification of × 1000. Random grain counts were made using an ocular grid of 400 squares, with a total area of 0.01 sq.mm. This grid was also used to measure the dimension of the injection site.
(B) Afferents Sources of afferents of the NA were examined using the Fink-Heimer I and Nauta-Gygax methods of labeling degenerating axonsg,zT. Cases were selected to examine degeneration within the NA following lesions in the septum (10 cases), cingulate gyrus (8 cases), hippocampus (3 cases), orbital frontal cortex (2 cases), ventral nucleus of the diagonal band (1 case), and caudate nucleus (1 case). The nomenclature of the septal nuclei has been adopted from Stephan and Andy4L For silver impregnation studies, either electrolytic lesions or fornix transections (scissor cut using offset spear scissors mounted on an electrode carrier) were placed stereotaxicallyzl. Following a 7-day postoperative period, the animals were perfused as previously stated. Later, frozen coronal sections (25 #m) were cut and stained. RESULTS
The nucleus accumbens (NA) is a large group of cells extending from the bed nucleus of the stria terminalis to levels anterior to the genu of the corpus caUosum and the anterior olfactory nucleus. The nucleus lies ventral to the anterior horn of the lateral ventricle between the diagonal band and the arm of the anterior limb of the anterior commissure. There appear to be two components to the nucleus. The lateral part lies between the ventral tip of the anterior horn of the lateral ventricle and the putamen. The cells are interspersed with transversing fibers and are somewhat scattered. The medial more cellular part extends from the ventral tip of the anterior horn of the lateral ventricle to the island of Calleja and the diagonal band. This medial part is the primary area of this study.
(A) Efferents In two of the animals the injection of labeled leucine was reasonably confined to the limits of the NA. The data indicating terminals in target nuclei for these animals were nearly identical. One of the injection sites is illustrated in Fig. 1. During the injection of radioactive leucine, some of the amino acid would likely be conveyed outward from the center of the injection site by simple diffusion. Previous studies have shown that labeled amino acid is preferentially absorbed by cell bodies and is not taken up by axons to any great extent 7. Thus, the leucine injected for this study was absorbed by perikarya around the injection center. In the prepared slides, labeled leucine is made to appear as silver grains representing amino acid bound as protein. Moreover, the absorption time of the labeled leucine is sufficiently short to indicate that most or all of the injectate was absorbed by perikarya at the injection site and was not later conveyed to other areas of the brain. Within the injection site the majority of perikarya are covered with silver grains
392
A Fig. 1. Illustrations of injection site for radioactive leucine. The stippling in A indicates the location of the injection in the nucleus accumbens for S-76. The center photograph indicates the area of symmetrical diffusion at the injection site. C, taken from the inset in B, shows an intense grain density over neuronal somata as well as between neurons. For abbreviations used in this and followingfigures see p. 401.
(Fig. 1C). Outside each of the blackened cell bodies, whorled tracts of grains are visible. These tracts may overlie proximal axons, and possibly dendrites, packed with radioactive proteins. As one moves from the center of the injection site toward the periphery, the number of blackened cells decreases as the concentration of grains overlying cell bodies and neuropil diminishes. The amount of labeling over perikarya decreases until the concentration of grains over them equals that observed over the surrounding neuropil. We have considered this to be the end of the diffusion area at the site of injection. We have defined 'diffusion space' as that area in and around the injection site in which perikaryal labeling equals or exceeds neuropil labeling. The diffusion space illustrated in Fig. 1 extended from the ventral tip of the right lateral ventricle 3.35 mm to the ventral border of the brain, possibly including a portion of the olfactory tubercle; from the isle of Calleja, it extended laterally 1.48 mm; in an antero-posterior dimension, the diffusion space extended 1.60 mm (A-P 15.3 to A-P 13.7). This shape was largely due to the slant of the needle tip which made the opening ovoid. The needle tract also may have permitted some longitudinal spread along its penetration. The volume of this ellipsoida| diffusion space was calculated to be 7.54 #1. The total number of cells within the diffusion space of the nucleus accumbens, calculated to be available to absorb the radioactive leucine, was 190,000 cells. Terminal projections of the NA, as revealed by the autoradiographic procedures, are illustrated for several areas in Figs. 2 and 3. The greatest density of grains was found in the lateral preoptic area. The grain pattern in the lateral external nucleus of the septum and lateral hypothalamus had the next highest density (Fig. 3). Grains
393
/
W-, /
j
f
.-,!~..... •.
~
i/////J~
~~.,~7 ',I 0 ' ~
~4 ~
~,
.....
/
F@ ,=L Fig. 2. Pattern of silver grains representative of terminal endings, following injection of radioactive leucine into the nucleus accumbens (A). The intense stippling in the nucleus accumbens illustrates the injection site for S-75. This section is 0.5 mm posterior to the center of the injection. All other stippling (grain patterns) in the illustrations from A to F represent terminals. B is 2 mm posterior to A.
were f o u n d t h r o u g h o u t the septal nuclei. L a b e l i n g was o b s e r v e d in the a n t e r i o r h y p o t h a l a m u s including the p a r a v e n t r i c u l a r a n d s u p r a c h i a s m a t i c nuclei. The lateral a n d d o r s a l h y p o t h a l a m u s c o n t a i n e d labeling as far p o s t e r i o r as the m a m m i l l a r y b o d y . G r a i n s were o b s e r v e d in the s u b s t a n t i a i n n o m i n a t a as far l a t e r a l as the e n t o p e d u n c u l a r nucleus b u t were n o t o b s e r v e d to be a b o v e b a c k g r o u n d levels m o r e l a t e r a d or in the a m y g d a l a . A restricted p a r t o f a r e a 24 o f the cingulate gyrus a n d the a n t e r i o r tip o f the h i p p o c a m p u s n e a r the uncus also c o n t a i n e d a n u m b e r o f grains. G r a i n s i n d i c a t i n g sites o f a x o n terminals were f o u n d to be o f a m o r e restrictive p a t t e r n in the t h a l a m u s ,
394
Fig. 3. Pictures of target structures demonstrating the density of silver grains present after a 12-week exposure period. Background activity was not significantly increased during this exposure time as compared with sections exposed for 6 weeks. However, the target areas were more easily distinguished at 12 weeks of exposure. Hippocampal gyrus (Hg) represents a control, i.e., background level of grain density at 12 weeks.
395
Fig. 4. Additional controls, i.e., examples of the density of background activity as represented by grain concentration at 12 weeks of exposure (S-76). NA, contralateral nucleus accumbens; Cg, ipsilateral cingulate gyrus area 23; Pu, ipsilateral putamen; Fa, ipsilateral fastigial nucleus. Thalamic nuclei over which grain counts exceeded that of control areas were the parataenial, periventricular, reuniens and dorsal medial. The habenular nuclei were the only other diencephalic structures to show labeling. The caudate nucleus (the medial part of the head and anterior part of the body) and pre-rubral field were the heaviest labeled extrapyramidal areas. Grains observed over the pars compacta of the substantia nigra were slightly less than those found in the thalamus. The globus pallidus had little or no labeling in the sections studied.
396
Fig. 5. Composite illustration of silver impregnation data showing representative lesions (top row across); subsequent argyrophilic pattern in the N A for each type of lesion (center row), and photomicrographs of the silver impregnated fibers for the cases used (bottom row). The pictures in the two left frames were taken from sections using the Fink-Heimer I technique for cases S-63 and S-38. The picture in the bottom right corner was taken from a section stained using Nauta-Gygax technique, case S-22. The calibration line applies only to the 3 photomicrograhs. 0.7 centimeter equals approximately 1 mm in the diagrams.
397 Grains were not above background activity in the amygdala, contralateral NA, fastigial nucleus and thalamic nuclei other than those reported above. Examples of control areas and background counts are shown in Fig. 4.
(B) Afferents The argyrophilia observed in the NA following lesions in the septum, cingulate gyrus and diagonal band differed in amount and in pattern of distribution. Following electrolytic lesions of the septum, argyrophilia was consistently observed within the NA (Fig. 5). The lesions were located dorsal to the anterior commissure and within the near margin of the corpus callosum from A13 to A 15. The electrode tract penetrated the overlying cingulate gyrus and superior frontal gyrus. Some component of the dorsal septal nuclei was involved in each case, i.e., the anterior, external, internal or intermediate parts. The degeneration patterns in the NA for these cases did not differ topographically from one another. The degeneration pattern was strongest in the medial and ventral portions of NA. Lesions of the cingulate gyrus were located from A8 to A10 in Brodmann's area 24. The lesion illustrated was small and largely confined to the cingulate gyrus. Other cases sustained larger lesions involving damage to the overlying frontal gyrus. However, greater amounts of argyrophilia were not observed within the NA for these latter cases. Therefore, the more anatomically confined lesion is shown in Fig. 5. Degeneration was present in the central area and on the medial edge of the nucleus. The overall density of argyrophilia was slightly less than it was following the septal lesion. The lesion in one animal was located in the diagonal band near the olfactory tubercle area (A14) but did not involve the medial part of the NA (Fig. 5). Heavy impregnation of fibers was seen on the medial edge of the NA (Fig. 5). The argyrophilic pattern was restricted to a narrow band along the medial edge and ventral border of the NA. Degeneration was found in the NA following a lesion of the medial part of the head of the caudate nucleus. This lesion had septal involvement, and the density or pattern of degeneration in the nucleus accumbens was not significantly different from that observed following lesions confined to the septum. The NA contained no argyrophilic particles following transection of the fornix, small lesions of the anterior cingulate gyrus, or lesions of the orbital-frontal cortex. The fornix of three animals was unilaterally sectioned (A7, L2.5, H + 9) using a stereotaxically placed extended scissors. The transection appeared to be complete in one case. The second animal's transection was similar but the degeneration in the lateral tip was less intense. In the third case, the lateral tip was free of degeneration. A large amount of argyrophilia was observed in the fornix column, anterior nuclei of the thalamus and mammillary body. No evidence of degeneration was seen in the NA in sections of any of the 3 animals although degeneration was obvious in main target structures in the thalamus (anterior nuclei, dorsal medial nucleus and dorsal lateral nucleus) and hypothalamus, especially the mammillary nuclei. Lesions of the anterior cortex were made in 4 animals by suction. All involved
398 the frontal gyrus with encroachment of the dorsal one-fourth of the cingulate gyrus, Brodmann's area 24 (A!8-19). A lesion was made by suction in the anterior half of the orbital-frontal cortex of two animals. Equal amounts of the overlying frontal lobe were also removed to gain access. Fiber and preterminal labeling were obvious in the internal capsule and dorsomedial nucleus of the thalamus. However, none was observed in the NA from either case. DISCUSSION
The results show that the NA is connected to the lateral hypothalamus, other limbic structures and extrapyramidal structures. The location of the NA would allow it to be related morphologically to either the caudate or septal nuclei. However, most of the results demonstrate that the connections of the NA are predominately within the hypothalamus and limbic system.
(A) Efferents Various investigators have described both slow and fast components of axoplasmic transport of incorporated radioactive leucine 7,za. Data of the efferent connections of this study are based upon the fast component. If one assumes that all of the free amino acid is taken up before the injection needle is removed, little if any labeled material would follow the tract during and after its withdrawal. The total amount of labeled leucine injected was 0.2 nmoles, administered in 0.04 nmole doses every minute for 5 rain. Five additional minutes were allowed before withdrawal of the needle. Assuming that all of the leucine transport proteins were maximally operative during the entire injection, the rate of transport of this amino acid across the neuron membrane equalled 0.9 nmoles transported per minute or the absorption time would be 15 sec. If non-saturated conditions are assumed, approximately 11 min would be required for absorption of the total injection of labeled leucine. The pulsed nature of the injection favors maximum velocity of uptake; therefore, the time span allotted is sufficiently long to expect that all of the labeled leucine was absorbed prior to removal of the needle. This would minimize the possibility of labeled tracer creeping along the needle tract and inadvertently labeling additional structures. This rate of uptake also strengthens the point that grain counts above background levels in nuclei beyond the bounds of the diffusion space (7.54/A) are terminal sites which have accumulated the labeled tracer. However, we were not able to differentiate between possible diffusion of isotope and what we believed to be short axon projections. Neuronal injury at the site of injection may have occurred as a result of introducing the 26-gauge needle into the N A tissue. The number of cells injured compared to the total number of cells in the diffusion space was so small that even if all the cells over the width of the needle tract died (approx. 8000) it would not decrease the total volume sufficiently to negate the validity of the above calculations. Our data show that the lateral hypothalamus is one of the major target sites of
399 the NA. Various investigators 11,19,a7 have reported this connection and described the pathway as coursing through the medial forebrain bundle. The results indicate that projections of the NA are topographic to the anterior cingulate gyrus. Powel130 traced degeneration to this portion of the cingulate gyrus in the rat after placing a lesion in the lateral septal nuclei which did overlapping damage to the NA. The results indicate a general projection to the septal nuclei. This is in partial agreement with studies using lesions which had slight encroachment on the NA19,22, 44. These workers reported degeneration within a limited septal region. Our injections, while directed to the medial half of the NA, were largely confined to the NA and occupied a large part of the nucleus. Connections from the NA were found to be distributed to the diagonal band within the olfactory tubercle and to structures within the diencephalon, i.e., thalamic midline nuclei (parataenial, periventricular, reuniens and dorsal medial) and the medial habenular nuclei. Authorsla, 19 have reported a substantial connection from the NA, to the medial part of the dorsal medial nucleus. We believe that the extra amount of degeneration present in their work was probably due to surgical involvement of the septum or more likely to fiber paths from the orbital-frontal cortex 31. The relatively high counts in the medial habenular nucleus appear to arise in the NA since the olfactory tubercle region projects preferentially to the lateral nucleus 28. Our data on non-limbic targets of the NA support the work of others who have also reported target nuclei to be the caudate ~2 and substantia nigra a7,41. However, we were unable to confirm connections with the globus pallidus as reported by Smith 41. (B) Afferents
Most of the afferent connections of the medial compact portion of the NA come from the septum (a limbic structure). This is in agreement with data reported by several authorsa0,z6, 3s. The fibers appear to originate in the septum and not from fornix fibers traversing this region, since no degeneration was found in the NA following transection of the fornix. Some authorsl,3,10,19 have reported degeneration in the NA after fornix section, but extensive surgical damage was shown in their studies. In a recent study using Fink-Heimer I silver impregnation methods, in the monkey, Seigel et al. ~9 observed terminations in the NA following hippocampal lesions using a 2-day postoperative survival period. It is possible that the terminals are structured in such a way (size?) that they are not visible after a 7-day postoperative period as used in our study. Also, in spite of the intensity of degeneration in the fornix column, thalamus and hypothalamus, it is possible that the lateral tip of the fornix was spared in all 3 animals in our transections of the fornix. Seigel et al. 39 reported that the hippocampus projected to the NA via the lateral fibers of the fornix. The cingulate gyrus (limbic cortex) had the next largest afferent connection to the medial part of the NA. This connection demonstrated a definite topography since only those cingulate lesions within a restrictive part of area 24 produced degeneration in the NA. The rather limited more anterior lesions of the cingulate gyrus were
400 found to give negative results. Krieg 20 states that area 24 of the cingulate gyrus probably has a variety of specific topography in the majority of its subcortical projections. Such highly topographic relations may be the reason why Domesick 6 and Powell et al. 34 did not report this connection. The diagonal band part of the olfactory tubercle, another timbic region, was shown to project to the NA. A few authors 2~,30,44 have reported this connection to be more extensive than our results indicate. The quantitative representation of this connection in our study may appear to be deficient because of the staining limitations of the N a u t a - G y g a x method, i.e., fine terminals do not stain well 32. The results indicate that the anterior half of the orbital frontal cortex does not project to the NA. More posterior areas of the gyrus were avoided because of the juxtaposition of the NA. Studies using silver impregnation methods 19 and electrophysiological data (evoked potentials, 4 msec latency) indicate that the amygdala may be monosynaptically related to the nucleus accumbens 33. This is in agreement with data (antidromic responses) collected from amygdaloid units following stimulation of the nucleus accumbens 16. Other origins of the N A afferents have been reported by various workecs. Projeztions to the N A were found to arise from thalamic structures, e.g., dorsal medial nucl~us (limbic) 23, parataenial nucleus2,35, 40 and parafascicular nucleus 4°. Guillery 12 reported finding projections to the NA from the hypothalamus. Connections of the N A are not limited to limbic structures; however, these few connections (non-limbic) occurred with extrapyramidal structures. Our work on the NA afferents had only one lesion in a non-limbic structure, i.e., the caudate nucleus. This lesion had involvement of the septum, and therefore the origin of the degeneration could not be determined precisely. K n o o k 19 also had the problem of septal involvement and was inconclusive about caudatoaccumbens connections. Other afferents were reported from substantia nigra 2~ and fastigial nucleus of the cerebellum 2s. CONCLUSION
The combined afferent and efferent data are summarized in Fig. 6. Our data indicate that most interconnections are between limbic structures and the NA. There is a cleat and extensive connection with the lateral hypothalamus and the septum. O f the non-limbic structures discussed, all except the cerebellum are considered to be within the extrapyramidal system. The extrapyramidal system and cerebellum are both functionally related to somatic muscle control. Data from recent experimentation suggest to us that the N A is involved in motor function 29 and may even differentiate in conjunction with motor demands 4. Considering that the connections are with limbic and extrapyramidal structures, the NA may serve as a bridge between limbic and extrapyramidal m o t o r systems effecting limbic influence in motor behavior. Eating and chewing, for example, are mixed visceral and m o t o r activities typical of limbic system involvement in phenomena like hunger, thirst, and fight behaviors. The
401
Fig. 6. S u m m a r y of connections of the N A . Solid lines represent efferent connections based on the autoradiographic data of this study. The dotted lines represent the afferent connections based on the data obtained f r o m the silver impregnation studies. The dashed lines represent afferent connections described by other authors.
muscles controlling jaw movement are phylogenetically and ontogenetically derived from the pharyngeal arches. Furthermore, temporal lobe seizure activity commonly involves jaw movement and salivation which are primarily visceromotor manifestations. LIST OF ABBREVIATIONS USED Ac = AmAv = Cc = Cd = Cg = Cp = Db = Dh = Dm = Ds = Fa = Fx = Gp = Hg = HI = Hm =
anterior commissure amygdala anteroventral nucleus of the thalamus corpus callosum caudate cingulate gyrus cerebral peduncle diagonal band dorsal h y p o t h a l a m u s dorsal medial nucleus o f the thalamus dorsal septal nucleus fastigial nucleus fornix globus pallidus hippocampal gyrus lateral habenula medial habenula
Hp Ic Ig Ld Lh Ls Mb Mh Ms NA Oc Of Op Ot Pa Pr
= = = = = = = = = = = = = = =
hippocampus internal capsule indusium griseum lateral dorsal nucleus of the thalamus lateral h y p o t h a l a m u s lateral septum mammillary body medial h y p o t h a l a m u s medial septal nucleus nucleus accumbens optic chiasm orbital frontal cortex optic tract olfactory tubercle paraventricular nucleus of the hypothalamus = pre-rubral field
402 Pt parataenial nucleus of the thalamus Po preoptic area Pu - putamen Pv - periventricular nucleus of the thalamus Sc suprachiasmatic nucleus of the hypothalamus Se :, septum Sn substantia nigra Sm stria medullaris thalami
Ti Th Un V Va
temporal lobe thalamus uncus ~ ventricle ventral anterior nucleus of the thaiamus Vh .... ventral hypothalamus Zi zona incerta
ACKNOWLEDGEMENTS T h i s s t u d y was s u p p o r t e d by N S F G r a n t G B - 3 2 1 7 0 . T h e a u t h o r s wish to t h a n k D r . S. A. G i l m o r e f o r h e r h e l p w i t h t h e l a b o r a t o r y p r o c e d u r e s f o r a u t o r a d i o g r a p h y , Dr. D o n D e L u c a f o r his assistance w i t h c o m p u t a t i o n s a n d P a u l R o b i n s o n for his t e c h n i c a l aid.
REFERENCES 1 CARMAN,J. B., COWAN, W. M., AND POWELL, T. P. S., The organization of cortico-striate connexions in the rabbit, Brain, 86 (1963) 525-562. 2 COWAN, W. M., AND POWELL, T. P. S., The projection of midline and intralaminar nuclei of the thalamus of the rabbit, J. NeuroL Neurosurg. Psychiat., 18 (1955) 266-279. 3 CRAGG, B. G., AND HAMLYN, L. H., Histological connections and visceral actions of components of the fimbria in the rabbit, Exp. Neurol., 2 (1960) 581-597. 4 CREPES, E. S., Time of origins in preoptic and septal areas of the mouse: an autoradiographic study, J. comp. NeuroL, 157 (1974) 161-244. 5 CROSBY, E. G., HUMPHREY, T., AND LAUER, E. W., Correlative Anatomy of the Nervous System, Macmillan, New York, 1962, pp. 360 and 419. 6 DOMESICK, V. B., Projections from the cingulate cortex in the rat, Brain Research, 12 (1969) 296-320. 7 EDWARDS,S. B., The ascending and descending projections of the red nucleus in the cat: an experimental study using an autoradiographic tracing method, Brain Research, 48 (1972) 45-63. 8 EMMERS, R., AND AKERT, K., A Stereotaxic Atlas of the Brain of the Squirrel Monkey (Sahnini sciureus), University of Wisconsin Press, Madison, Wisc., 1963, pp. 3 ft. 9 FINK, R. P., AND HEIMER, L., Two methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system, Brain Research, 4 (1967) 369-374. 10 Fox, C. A., The stria terminalis, longitudinal association bundle and precommissural fornix fibers in the cat, J. comp. NeuroL, 79 (1943) 277-296. 11 GURDJIAN,E. S., The corpus striatum of the rat, J. comp. NeuraL, 45 (1928) 249-281. 12 GUILLERY,R. W., Degeneration in the hypothalamicconnexion of the albino rat, J. Anat. (Land.), 91 (1957) 91-115. 13 GUILLERY,R. W., Afferent fibers to the dorso-medial thalamic nucleus in the cat, J. Anat. (Lond.), 93 (1959) 403~,19. 14 HEATH, R. G., Physiologic basis of emotional expression: evoked potential and mirror focus studies in rhesus monkeys, Biol. Psychiat., 5 (1972) 15-31. 15 HENDRICKSON,A., MOE, L., AND NOBLE, B., Staining for autoradiography of the central nervous system, Stain Technol., 47 (1972) 283-290. 16 ITO, N., ISHIDA, H., M1GAKAWA,F., AND NAITO, H., Microelectrode study of projections from the amygdaloid complex to the nucleus accumbens in the cat, Brain Research, 67 (1974) 338-341. 17 JOHNSTON,J. B., Further contributions to the study of the evolution of the forebrain, J. comp. NeuroL, 35 (1923) 337-382. 18 KAPPERS, C. U., UND THEUNISSEN,W. F., Die Phylogenese des Rhinencephalons, des Corpus striatum, und der Vorderhirnkommissuren, Folia neuro-biol. (Lpz.), 1 (1970) 173-288.
403 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
KNOOK, H. L., The Fibre Connections of the Forebrain, Davis, Philadelphia, Pa., 1966, 145 pp. KREIG, W. J. S., Connections of the Cerebral Cortex, Brain Books, Evanston, II1., 1963, 219 pp. LAS~K, R. J., Protein transport in neurons, Int. Rev. Neurobiol., 13 (1970) 289-324. LAUER, E. W., The nuclear pattern and fiber connections of certain basal telencephalic centers in the macaque, J. comp. Neurol., 82 (1945) 215-254. LEONARD, C. M., The prefrontal cortex in the rat. (I) Cortical projection of the mediodorsal nucleus. (II) Efferent connections, Brain Research, 12 (1969) 321-343. LORENS,S. A., SORENSEN,J. P., ANO HARVEY,J. A., Lesions in the nuclei accumbens septi of the rat: behavioral and neurochemical effects, 3. comp. physioL PsychoL, 73 (1970) 284-290. MOORE,R. Y., BHATNAOAR,R. K., AND HEELER, A., Anatomical and chemical studies of a nigroneostriatal projection in the cat, Brain Research, 30 (1971) 119-135. NAUTA, W. J. H., Hippocampal projections and related neural pathways to the mid-brain in the cat, Brain, 81 (1958) 319-340. NAUTA, W. J. H., AND GYGAX, P. A., Silver impregnation of degenerating axons in the central nervous system. A modified technique, Stain Technol., 92 (1954) 91-93. PAUL, S. M., HEATH, R. G., AND ELLISON,J. P., Histochemical demonstration of a direct pathway from the fastigial nucleus to the septal region, Exp. Neurol., 40 (1973) 798-805. PLINENBURG,A. J. J., WOODRUFF,G. N., AND VAN ROgSUM,J. M., Ergometrine induced locomotor activity following intracerebral injection into NAS, Brain Research, 59 (1973) 289-302. POWELL,E. W., Septal efferents revealed by axonal degeneration in the rat, Exp. Neurol., 8 (1973) 406-422. POWELL, E. W., Limbic projections to the thalamus, Exp. Brain Res., 17 (1973) 394-401. POWELL, E. W., AND SCHNURR,R., Silver impregnation of degenerating axons; comparisons of postoperative intervals, fixatives and staining methods, Stain TechnoL, 47 (1972) 95-100. POWELL, E. W., CLARK, W. M., AND MAUKAWA,J., An evoked potential study of limbic projections to nuclei of the cat septum, Electroenceph. clin. Neurophysiol., 25 (1968) 266-273. POWELL,E. W., AKAGI, K., AND HATTON,J. B., Subcortical projection of the cingulate gyrus in the cat, J. Hirnforsch., 15 (1974) 313-322. POWELL, T. P. S., AND COWAN, W. M., A study of thalamostriate relations in the monkey, Brain, 79 (1956) 364-390. RAISMAN,G., The connexions of the septum, Brain, 89 (1966) 317-368. RIOCH, D. M., Certain myelinated fiber connections of the diencephalon of the dog, cat and aevisa, J. comp. NeuroL, 53 (1931) 319-388. SEIGEL, A., AND TASSONI, J. P., Differential efferent projections of the lateral and medial septal nuclei to the hippocampus in the cat, Brain Behav. Evolut., 4 (1971) 201-219. SEIGEL, A., OHGAMI, S., AND EDINGER, O. i . , Projections of the hippocampus in the squirrel monkey, Presented at Society of Neuroscience 4th Annual Meeting, (1974). SIMMA,K., g u t Projektion des Centrum medianum und Nucleus parafascicularis thalami beim Menschen, Mschr. Psychiat. NeuroL, 122 (1951) 32--46. SMITH,O. C., The corpus striatum, amygdala, and stria terminalis of Tamandua tetradactyla, J. comp. NeuroL, 51 (1930) 65-127. STEPHAN, H., AND ANDY, O. J., Cytoarchitectonics of the septai nuclei in old world monkeys (Cercopithecus and Colobus), J. Hirnforsch., 7 (1964) 1-23. TUCHMANN-DUPLESSIS,H., AUROUX,i . , AND HAEGEL,P., Illustrated Human Embryology, Vol. 3, Nervous System and Endocrine Glands, Springer, New York, 1974, pp. 60-71. WOODS,W. H., HOLLAND,R. C., AND POWELL, E. W., Connections of cerebral structures functioning in neurohypophysial hormone release, Brain Research, 12 (1969) 26-46.
