Cortical projection patterns of the medial septum-diagonal band complex

THE JOURNAL OF COMPARATIVE NEUROLOGY 293103-124 (1990)

Cortical Projection Patterns of the Medial Septum-Diagonal Band Complex

RONALD P.A. GAYKEMA, PAUL G.M. LUTTEN, CSABA TWAKAS, AND JORG TRABER Department of Animal Physiology, University of Groningen, Haren, The Netherlands

(R.P.A.G., P.G.M.L., C.N.), and Neurobiology Department, Troponwerke, Cologne, FRG (J.T.)

ABSTRACT A detailed analysis of the cortical projechns of the medial septum-diag-

onal band (MS/DR) complex was carried out by means of anterograde transport of Phaseolus vrilgaris leucoagglutinin (PHA-L). The tracer was injected iontophoretically into cell groups of the medial sept,um (MS) and the vertical and horizontal limbs of the diagonal band of Broca (VDR and HDB), and sections were processed immunohistochemically for the intra-axonally transported PHA-L. The labeled efferents showed remarkable differences in regional distribution in the cortical mantle dependent on the position of the injection site in the MSiDB complex, revealing a topographic organization of‘ the MSIDB-cortical projection. In brief, the lateral and intermediate aspects of the HDR, also referred to as the magnoc:cllular preoptic area, predominant.ly project to the olfactory nuclei and the lateral entorhinal cortex. The medial part of the HDR and adjacent caudal (angular) part of the VDB are characterized by widespread, abundant projections to medial mesolimbic, occipital, and lateral entorhinal cortices, olfactory bulb, and dorsal aspects of the subicular and hippocampal areas. Projections from the rostromedial part of the VDB and from the MS are preponderantly aimed at the entire hippocampal and retrohippocampal regions and t,o a lesser degree a t the medial mesolirnbic cortex. Furthermore, the MS projections are subject to a clear mediolateral topographic arrangement, such that the lateral MS predominantly projects t,o the ventral/temporal aspects of the subicular complex and hippocampus and to the medial portion ofthe entorhinal cortex, whereas more medially located cells in the MS innervate more septal/dorsal parts of the hippocampal and subicular areas and more lat,eral parts of the entorhinal cortex. PHA-L filled axons have been observed to course through a number of pathways, i.e., the fimbria-fornix system, supracallosal stria, olfactory peduncle, and lateral piriform route (the latter two mainly by the HDB and caudal VDB). Generally, labeled projections were distributed throughout all cortical layers, although clear patterns of lamination were present in several target areas. The richly branching fibers were abundantly provided with both “boutons en passant” and terminal boutons. Both distribution and morphology of the labeled basal forebrain efferents in the prefrontal, cingulate, and occipital cortices closely resemble the distribution and morphology of the cholinergic innervation as revealed by immunohistochemical demonstration of choline acetyltransferase. In contrast, the labeled projections to the olfactory, hippocampal, subicular, and entorhinal areas showed a heterogeneous morphology. Here: the distribution of only the thin varicose projections resembled the distribution of cholinergic fibers.

Accepted September 18,1989. Address reprint reqrirsls to R.P.A. Gaykema, Department of Animal Phys-

iology, University of Groningen, P.O. Ron 14, 9750 AA Haren. The Nether- lands.

0 1990 WILEY-LISS, INC.

104 R.P.A. GAYKEMA ET AL.

Key words: basal forebrain cholinergic system, cerebral cortex, horizontal and vertical limbs of the diagonal band, anterograde tracing, Phaseolus vulgaris leucoagglutinin

In recent years considerable knowledge has accumulated on the extrinsic source of input to the olfactory, hippocampal, amygdaloid, and cortical regions in the mammalian brain, originating in the nuclei o f the medial septum (MS), the vertical and horizontal diagonal hand of Rroca (VDB and HDB), and the magnocellular basal nucleus (MBN). The predominant cholinergic nature of this projection has clearly been indicated by preliminary studies applying lesions in the basal forebrain, resulting in subsequent loss of cholinergic marker enzymes in the above-mentioned target areas (Lewis and Shute, '67; Shute and Lewis, '67; Mellgren and Srebro, '75; Wenk et al., '80; Johnston et al., '81). These early findings received further support from studies in which retrograde transport, methods were combined with histochemical detection of acetylcholinesterase (AChE) (Henderson, '81; Macrides e t al., '81; Big1 e t al., '82; McKin- ney e t al., '83; Mesiilam e t al., '83a,b; Alonso and Kiihler, '84) or with simultaneous immunocytochemical detection of choline acet,yltransferase (ChAT) (Rye et al., '84; Woolf et al., '84; Amaral and Kurz, '85; Zaborsky et al., '86). interest in t.he basal forebrain cholinergic system (BFChS) has greatly increased since neuropathological studies in humans have provided evidence that the BFChS is consist.ently and severely affected in Alzheimer-type dementia (Rartus et al.. '82; Coyle et al., '83; Whitehouse et al., '83; Arendt e t al., '85; Saper et al., '85). Moreover, animal experiments show that cortical cholinergic neurotxansmission plays an integral role in learning and memory processes (Flicker et al., '83; Spencer and Lal, '83).

The various studies employing r e h g r a d e tracing techniques mentioned above have provided a global image of the topographic relations between the cholinergic basal forebrain nuclei and their cortical target structures. It has become clear that the MS and the VDB are the major sources of cholinergic input to the hippocampus, whereas t,he HDR provides efferents to the olfactory bulh, although the MS-, VDB-, and HDB projections to the hippocampal, cntorhinal, and olfactory areas also include a considerable noncholinergic component, (Alonso and Kohler, '84; Rye et al., '84; Woolf et al., '84; Amaral and Kurz, '85; ZBborsky et al., '86; Freund and Ant.al, '88). The MBN has been shown to be the subcortical source of cholinergic input to the neocortex, for the greatest part being cholinergic in nature (Mesu- lam et al., '83b; Rye et al.? '84).

Our current knowledge of the distribution of the cholinergir innervation in the cortical mantle is largely based on the demonstration of AChE- and ChAT-posit.ive fibers (Storm- Mathisen and Blackstad, '64; Macrides et al., '81; Mesulam et al., '84; Wainer et al., '84b; Frotscher and LCranth, '85; I-louser et al., '85; Hellendall e t al.? '86; Lysakowski et al., '86, '89; Parnavelas et al., '86; Brady and Vaughn, '88; Eck- enstein et, al., '88). To date, the relationship between the cortical cholinergic fibers and the cells of origin in the basal forebrain, however, has not. been fully understood. This is partly due to the fact that the cortical ChAT-positive fibers represent hoth extrinsic and intrinsic sources of cholinergic innervation of the cortex and hippocampus (Houser et al., '83, '85; Wainer e t al., '85; Frotscher et al., '86; Parnavelas et al., '86; Matthews e t al., '87; Blaker et, al., '88; Eckenstein e t al., '88). Furthermore, our knowledge on the topographic

arrangement of t.he basal forebrain cholinergic projection to the cortex is largely based on retrograde transport studies. Retrograde tracing methods do not provide data on morphology and distribution patterns of the cortical afferents from the basal forebrain. By using anterograde tracing techniques, the pattern of the cortical innervation that origi- nates in t,he basal forebrain can be visualized and related to the cells of origin in the various subdivisions of the basal forebrain. In previous studies, some of such data was obtained with anterograde transport of labeled amino acids (Meibach and Siege], '77; Swanson and Cowan, '79; Milner et, al., '83; Saper, '84) or peroxidase-conjugated wheat germ agglutinin (Alonso and Kijhler, '84; Lamour et al., '84). How- ever. these studies did not provide a detailed survey of the terminal fiber morphology because of the limitations of the applied techniques, nor did they provide information on the fine topographic arrangement of the BFChS because of a relative limited number and large size of the tracer injections.

Therefore, the aim of the present study is to map in detail the terminal field of t,he basal forebrain efferents in the entire cortical mantle by employing anterograde intra- axonal tracing technique with Phascolus uulgaris leucoag- glut inin (PH.4-L). This technique visualizes a small number or neurons in a limited tracer uptake area, the trajectories of their individual fibers, collaterals, and the entire terminal synaptic fields a t light microscopic levels with a high resolu- tion (Gerfen and Sawchenko, '84; Ter Horst, et al., '84). As revealed by this method, we have described the efferent projections of the MBN to the cortex and the MS-VDB efferents to the hippocampus in a number of previous reports (Luiten et al., '85, '87; Spencer et al., '85; Nyakas et al., '87). In this work we report in detail on the projection patterns of the MS, VDB, and the HDB to the mesolimbic prefrontal and cingulate cort,ices, the entorhinal cortex, the subicular complex, and the occipital neocortex. The study is focussed on the regional cortical innervation patterns with reference to the precise origin of the PHA-L labeled projections in order t.o elucidate the fine topographic organization of the MS/I)R complex. Additional attention is given to the laminar distribution of the PHA-L laheled efferents as compared to t,he distribut,ion of the AChE- and ChAT-positive processes in the cortex.

MATERIALS AND METHODS PHA-L procedure

The aiiterograde tracing experimenk were carried oul on 62 male Wistar rat,s between 3 and 4 months of age. Each animal received a single, unilateral PHA-L iiijection in the region of the HDB, VDB, or MS. The animals were anesthetized with a combination of Hypnorm (Duphar, 0.4 mg/kg i.m.1 and sodium pentobarbital (30 mg/kg i.p.) and mounted into a stereotactic frame. PHA-L was delivered iontophoretically through bevelled glass micropipettes with tip diam- eters of 15 to 20 wm, which were positioned in the brain area defined hy the coordinate system of Paxinos and Watson ('86). The micropipet,tes were filled with a solution of 2.5:; PHA-1, (Vect,or Labs) in tris-buffered saline (TBS, pH 7.4)

105 SEPTAL-DIAGONAL BAND PROJECTIONS TO THE CORTEX

and connected to the positive pole of a constant current source (Midgard CS3). T h e current intensity for iontopho- retic delivery of the tracer varied between 5 and 6.5 p A and was applied for 30 45 minutes in a 7 -second o n h e c o n d off cvcle. After completion of the iontophoresis. the glass pi- pette was left in its posit.ion for I0 minutes to prevent leakage o f tracer in the track during retraction.

After recovery from the operation, the animals survived for 6 to 10 days. They were then deeply anesthetized with pentobarbital (70 mg/kg body weight i.p.) and the vascular bed perfused transcardially with a buff'ered mixture of 2.5 'I glut,araldehyde, 0.5 ''(< paraformaldehyde, and 4 ';. sucrose in 0.05 M phosphate buffer (pH 7.4). T h e fixed brains were stored overnight in buffered 30f'i, sucrose at 4"C, then rap- idly frozen 011 dry ice; 40-pm-thick coronal sections were cut on a cryostat microtome. The t.emporal hippocampal region of 5 hrains were cut in a horizontal plane. T h e sections were collected serially arid thoroughly rinsed in 0.05 M tris-buff- eted saline (TRS, 0.9"i, NaCl, pH 7.4). Every third section was processed for PHA-L irnmunoreactivity. T h e brain sections were subsequently incubated in goat ant,i-PHA-L (Vector, 1:2000,48 h), rabbit antigoat IgG (Sigma, l:200, 2-4 hours) a n d goat peroxidase-antiperoxidase complex (DAKO, 1:500,4 hours) a t 4°C in solutions containing 0.5 M NaC1, 0.5'( Triton X-100, and 0.05 M tris buffer (pH 8.6). T h e sections were then processed for 1 hour in a solution of 0.4 " I 3,3'-diaminohenzidine (Sigma) and 0.01 cr' H,O, in tris buffer a t pH 7.6, mounted, counterstained with cresylviolet, dehydrated, and coverslipped.

AChE-pharmacohistochemical procedure For mapping of AChE-positive neurons in t,he M S , VDB,

and HDB. a pharmacohistorhen~ical procedure as described by Butcher e t al. ('75) was carried out. Diisopropylfluoro- phosphate (DYP, Sigma) dissolved in arachis oil (1:1,000) was injected intramuscularly, at a dose of 1.5 mg/kg body weight, immediately followed by a n i.p. injection of 5 mg/kg atropinc. Aft,er 5 hours survival time, the above described PHA-L procedure started with perfusion, fixation, and cryosectioning of the brain. T h e Koelle copper-thiocholine procedure was applied to free-floating, 40-pm sections. Every third section containing the basal forebrain cholinergic nuclei, adjacent to one processed for PHA-L immunoreactivity, was selected for this purpose. The strongly AChE-stained neurons in the basal forehrain are generally also ChAT-immunopositive and thus can be considered as cholinergic (Eckenstein and Sofroniew, '83; Levey e t a]., '83a).

ACE-fiber staining Five adult male ra ts were used for demonstration of

AChE-positive fibers in the cortical mantle. 'l'hese animals were suhject t o the same perfusion and sectioning protocol as described above, except t h a t the brains were cut at a thickness of 10 and 20 pm. The sections were collected serially and stored in TBS prior t o processing for AChE according to a modified AChE-histochemical prcrcedure (Hedreen et al., -85). Sections were rinsed in two changes of

______~

ac Acb AChF, ACv,d ACO AId,v,p AOU AON UFChS ULA c.4 1-3

ChAT CPU df DG

RC EPl' f fi f mi fmJ G L G P GR Hb HDRm,li

I-tFd,v

IG IL LEA LO

c(

r)Lli.

I C

anterior commissure accurnbens nucleus avetylcholinesterase anterior cingulatr cortex, veritral and dorsal part. anterior rnrtiral ;imygdaloid nucleus agraniilar insular cortex, dorsal, ventral and poster accessory olfactory bulb anterior olfactory nucleus basal forebrain cholinergic system basolateral arnygdaloid niiclens subfields of the ammon's horn of the hippocampus corpus callosiim choline acetyltransferase caudate putamen dorsal fornix dentate gyrus dorsolateral subdivision of the entorhinal cortex entorhinal cortex external plexiforrn layer of the olfactory bulb fornix fimbria forceps minor of t.he curpus callosum lorceps major of the corpus callosurn glomerular layer of the olfactory bulb globus pallidns internal granular layer of the olfactm-y bulb habenular niirlens horizontal limb nf the diagonal band. medial and

hippocampal formatiori, dorsal and ventral part int.errial capsule induseum griseiim infralimbic cortex latcral entorhirial area lateral orbital cortex

l a t e r a l - i n t e ~ m e ~ ~ a t ~ , part

MBN M r)

lateral septa1 nucleus fmrrtoparietal motor cortex magnoccllular basal nucleus rnediodorsal nucleus of the thalamus medial subdivision of the rntorhinal cnrtex medial entorhinal arw medial amygdaluid nucleus rnrdial nrbital cortex medial septa1 nucleus, medial and lat,eral par t nucleus o l the lateral olfactory tract o l fx to ry bulb occipital cortex parasubiculnm medial precenrral cortex lateral precentral cortex piriform cortex prelimbic cortex posterolatrral cortical amygdaloid niicleus post.rromedia1 cortical arnygdaloid nucleus perirhinal cortex presrihi~wlurn retrosplenial cortex, granular and agranular part frontoparietal primary and seciindary sornatosensorp cortex stria nredullaris of the thalamus siihiculum twnporal cortex taenia tecta olfactory tubercle vertical limb of the diagonal band, rostromedial and

ventrolatcral subdivision of the entorhinal cortex vcntrolateral orbital cortex ventromedial subdivision of t he entorhinal cortex vmtr:il pallidurn

cii ii drrlateral par t

106

\ - /

R.P.A. GAYKEMA ET AL.

Fig. 1. Position and extent of the PHA-L injection sites (hatched areas) in the MWUB complex giving rise to projections to the cortical, hippocampal, and olfactory areas. The injection sites have been plotted in a series of transverse sections (A to G ) representing camera lucida

drawings from ChAT-immunostained sections of one brain for reference. The distribution of ChAT-positive neurons is indicated with black dots. Some control cases outside the MS/DB region with no cortical projections are encircled with broken lines.

0.1 M acetate buffer and subsequently incubated in a medium containing 0.05 '( acetylthiocholine-iodide, 3 mM cupric sulfate, 4 mM sodium citrate, and 0.1 mM potassium ferricyanide in 0.065 M acetate buffer (pH 6.0). Thereafter the staining was intensified by subsequent processings in a I ('( ammonium sulfide and a 0.1 7; silver nitrate solution.

CUT-immunohistochemistry Another 5 animals were studied for ChAT-immunoreac-

tivity. These animals were perfused with a buffered fixative solution containing 3'; paraformaldehyde in 0.05 M phosphate buffer (pH 7.4), followed by 100 ml of cold 10"~ sucrose in 0.1 M phosphate buffer (pH 7.4), and the frozen brains cut into 20- or 40-pm coronal sections. Sections were thoroughly rinsed in TBS and a selected number processed

for ChAT-immunoreactivity. The characterization, purifi- cation, and specificity of the monoclonal anti-ChAT antibody AR8 has been well documented (Levey et al., '83b; Wainer et al., 'Ma). Free-floating sections were rinsed in TBS for 1.5 minutes and rinsed subsequently in TBS between the incubations. All antibodies were diluted in TBS with 0..5', triton X-100 to which 1% normal goat serum wae also added. 'The sections were first incubated in 10% normal goat serum for 1 hour. Next, the sections were incubated in rat monoclonal anti-ChAT (]:loo, 48-72 hours, a t 4"C), followed by overnight incubations in goat antirat IgG (Zymed, 1:100) and in rat monoclonal peroxidase antiperoxidase complex (Sera-lab, 1:lOO). Finally, the sections were processed in DAB (0.04"0) and H,O, (0.01'' ) in tris-HC1 (pH 7.6), mounted, dehydrated, and coverslipped.

SEPTAL-DIAGONAL BAND PROJECTIONS TO THE CORTEX 107

TABLE 1. S u m m a r y of t h e intensi ty of anterograde labellng in the various cortical, olfactory, and hippucampal areas a f t e r injections in the HDR. VDB,

and MS.'

0 0

0 . 0 0 0 . 0 0 0

0 0

0 0

0 0 0 0 0 0 0 0 0 . 0

0 0 0..

*.. 0 0 0 0 0 . 0 . : 0 . 0 0 0 . 0 0 0 0 0 0 . 0 0 0 0 0 0

0 0 0 0 0 0 . 0 . . 0 . 0 0 .. 0 . . 0 . 0 0.0. 0 0 0 0 . 0 ........ 0 0 0 0 0 0.e.. ........ ..... 0 0 0 0

0 . 0

:oOt : t : : t 0 0

0 . 0 0 0 0 0 . 0 0 0 0 . 0 0 . 0 0 0 0 . 0 . 0 . 0 0 0 0 0 0 0 . 0 0 0 0 0 0 0 0 0 0 . . 0 0 0 . ........... 0 0 .............. ........... .. 0 0 . 0 0 0 . 0 0 . 0 . 0 . . 0 . 0 0 0 . . 0 . 0 . . 0 . . 0 0 0 0 .................. 0 0 ............. 0

0 0 0 0 .. 0 . . 0 . 0 . 0 . 0 0 0 0 0

0 0 . 0 0 0 . 0 0 0 . . 0 . . 0 0 . 0 0 . 0 . 0 0 0 . 0 0 . .... 0 . 0 . .

0 . 0 . . 0 0 . 0 . .

0 0

0.0.: ........... 0 . 0 . 0 . 0 . 0 . 0 . 0 0 . ........................

0 0 0 0 ................ 'Numbers on the top of the tahle refer to the experiment numbers and correspond to those in Figure 1. Experiments are listed from left to right reflecting a lateral to medial position within the HDB and MS and n caudolateral to roatromedial position within the VDB. Symbols: large filled circles: massive innervation, small filled circles: moderate innervation, open small circles: sparse innervation. Abbreviations: agr ins, agranular insular cortex; ant cg. anterior cingulate cortex; ant olf n, anterior olfactory nucleus; CA, f i~ lds C A 1 ~ 3 of lhe Ammon's horn; DG, dentate gyrue; dors, dorsal: ent, entorhinal cortex; fr par ms, frontoparietal somatosensory cortex; infralimb, infralimbic cortex; I, lat, laterla; med, medial; mot, motor cortex; occip, occipital cortex; olf bulb, olfactory bulb; orb, orbital cortex; parasub, parasubiculum; perirhin, perirhinal cortex; post, psterior; prec, pwcentral cortex; prelimb, prelimbic cortex; presub, presubiculum; retrospl aglgr, agranularlgranular retrosplenial cortex; sub, subiculum; temp, temporal cortex; ventr, ventral.

RESULTS Nomenclature of the cortex

In this report. a cortical topographic terminology is used for identification of cortical areas, which is based on several cytoarchitectonic studies (Krettek and Price, '77a,b; R.eep, '84; Zilles, '85). Specific attention is paid to (para-) limhic meso- and allocortical areas, since the majority of the MS/ DB efferents are aimed a t these regions of the cortical mantle. The medial mesocortex includes the prefrontal, cingulate, and retrosplenial regions, divided into medial orbital (MO), infralimbic (IL), prelimbic (PL), ventral and dorsal anterior cingulate (ACv,d), and granular and agranular retro-splenial cortex (RSg,a) (Krettek and Price, "77a; Zilles, '85). The lateral mesocortex comprises the orbital, agranular insular, and perirhinal regions. subdivided into ventrolateral (VLO) and lateral orbital (LO), dorsal, ventral, and posterior agranular insular (ATd,v,p) and perirhinal cortex (Prh) (Krettek and Price, '7Sa). The docor tex , as adapt.ed

from Iieep ('84), consists of the olfactory regions: main olfactory bulb (OB), anterior olfactory nucleus (AON), and piriform cortex (Pir). I t also includes the entorhinal cortex and the hippocampal formation. The entorhinal cortex is subdivided into dorsolateral (DLE). ventrolateral (VLE), ventromedial (VME), and medial parts (ME). The hippo- carnpal formation is divided in the pre- and parasubiculum (PrS and Pas) , subieulum (Sub), dentate gyrus (UG), and hippocampus proper (CA fields 1-4) (Krettek and Price, '$71~).

Nomenclature of the basal forebrain The MS/DB complex as a whole can easily he delineated,

but 1itt.le agreement exists about the topographic termincd- ogy of the complex subdivisions. Therefore, its nomenclature used here needs further explanation. The MS cholinergic neurons form a midline raphe and a substantial cell mass situated in the lateral half of the nucieus, separated by a contingent of noncholinergic neurons (Mesulam et al., '83b).


The M S borders the VDB ventral to the lateral mass of MS cholinergic neurons, where the cholinergic cells become sparse (Swanson and Cowan, '79). Posteriorly the VDB can be easily distinguished from the MS by the tubular arrangement o f the cholinergic cells, surrounding the fiber bundle of the VDB (Fig. 1C.D). Both in anterior and posterior directions, the VDB is more elongated than the MS, ventro- caudally approaching the medial border of the olfactory tu hercle.

The caudolateral part of the VDB (ZLborsky e t al., '86) or the pars angularis of the diagonal band (Rig1 et al.. '82; Fig. Ill) is referred to by some authors as the nucleus of the horizontal limb of the diagonal band (Swanson, '76; Woolf et al., '84). 'rhe HUB in this study represents the part of the complex situated deep to the caudal part of the olfactory tubercle and lateral to the medial preoptic area as designated by ZBbvrsky et al. ('86) (Fig. 1E-G). 'The dense population of large multipolar cholinergic neurons of the caudoventral edge of the VDB are thus continuous with those lateral to the medial prevptic area a t the level of the crossing of the anterior commissure, and in this work indicated as the medial part of the HDB (cf. Mesulam et al., '83b; Zaborsky et al., '86). The intermediate and lateral parts of the HDB, where a considerable lesser percentage of the neurons is cholinergic, have previously been referred to as the magnocellular preoptic region (Swanson, '76), but this term is not used here. Additional parcellation of the MS/DB complex, however, can be made with respect to projection patterns, as discussed below.

Survey of the PHA-L experiments Of the 62 technically successful PHA-L experiments, 21

cases produced FHA-L deposits that were centered beyond the houndaries of the MS/DB complex, as indicated by the distribution of the ChAT or AChE positive somata (Fig. I). The remaining 39 experiments were evaluated and analyzed for anterogradely labeled efferents in the various cortical target areas, of which 30 representative cases are summa- rized in Table 1. The size and location of the PHA-L injections are shown in Figure 1. In all experiments the delivery of the tracer resulted in labeling a small cluster of neurons of approximately 500 Fm in diameter (Fig. 2). The position of the injection sites and overlap with cholinergic cell bodies was determined by comparison of PHA-L deposits and AChE-stained neurons in alternate sections. All tracer injections covering areas rich in cholinergic somata resulted in anterograde labeling of projections to the cortical mantle (including hippocampus and olfactory bulb). Injections that did not coincide with the MS/I)B complex never gave rise to such projections. Some of the latter injection sites are drawn in Figure I for reference and are indicated by interrupted lines. The pattern of labeled fiber tracts originating from t,hese iionchoiinergic injection sites differed substantially from those cases! in which the PHA-L was centered within the MS/UB complex.

The results of the anterograde labeling of the efferents in the cerebral cortex following the PHA-L injections in the MS/DB complex are described on three subsequent levels. First, a general pattern of projections to the cortical mantle from the various subdivisions of the MS/DB complex is given, with emphasis on overall organization of the MS/DB Ijrojections to their target areas. Second, efferent tracing from the site of origin to the cortex is described in detail with special attention to the course of pathways and laminar arrangement of efferents. Third, the distribution and mor-

Fig. 2. Low-power photomicrographs illustrating the PHA-L injection sites in the lateral part of MS (a) and the lateral part of the HDB (b) for cases 150 and 67, respectively. Scale bar in (a) = 500 rcm, in (b) =

250 gum.

phology of the terminal innervation in the cortical target areas are surveyed a t a cellular level. The anterogradely labeled projections to the dentate gyrus and hippocampus proper have been described in detail elsewhere (Nyakas et al., '871, whereas the MS/DB efferents to the olfactory bulb are dealt with in a separate report. Abundant subcortical projectioiis to thalamus, hypothalamus, and midbrain appeared in all cases, but are not further considered here.

General pattern of the MS/DB projections to the cortical mantle

Overall pattern of the HDB projections to the cortex. The HDB as a whole can be divided into two subdivisions with respect, to the patterns of their projections to the cortex. In that sense a clear difference is present between the lat,eral and intermediate parts of the HDB (together referred to as HDBlii versus the medial part of the HDB (HDBm). HDB projections did not differ with respect to the anterior-posterior position of t.he tracer injection within the


Fig. 3. Chartings of a series of frontal sections through the rat telencephalon (A-H) illustrating the axon trajectories and projections to the olfactory, piriform, and entorhinal areas following a PHA-1, injection in the lateral part of the HDB (D).


Fig. 4. Chartings presenting a series of transverse sections through the rat brain (A-H) and illustrating the distribution of anterogradely labeled axons and terminals in the cortical, olfactory, amygdaloid, and

hippocampal The injection

following a pHA-L injection in the VDBc region. is located in D.


HDB. The main target struct,ures of the HDBli neurons are the olfactory regions (OR, AON, OT, and Pir), the ventral aspects of the taenia tecta as well as the medial and cortical arnygdaloid nuclei and the lateral entorhinal cortical areas (DJ,E and VLE) (Fig. 3) . In addition, the orbital and ant'e- rior insular (\'LO: LO. Alv, and AId) and the anterior medial mesocortical areas (MO and TL) receive some input from the HDBli. Compared to these mesolimbic projections, input, to the neocortex is more scarce and mainly directed to the medial precentral (PCm) and somatosensory cortical (SS) areas adjacent to the mesocortical regions.

The cortical projection pattern of the HDRm is entirely different from that of the HDBli and bears more resem- blance to the projection of the caudolateral parts of the VDR (Fig. 4). As such, the HDBm may be considered as a continuation of the VDR. The HDBm is characterized by an intense projection to the entire medial mesocortical region (MO. TI,, PL, ACd, ACv, RSg, and RSa)-often extending into the adjacent neocortical regions (PCm and motor area (M))-and to the occipital cortex (OC). The taenia tecla ('YI') and induseum griserim (IG) are also richly supplied with HUBm efferents. Furthermore, tracer injections in the HDHrri resulted in abundant terminal labeling in the dorsal aspects of the Sub, PrS, and the hippocampus. In addition, the HDRm maintains considerable projections to the olfactory regions (OB, AON, and Pir), the orbital prefrontal cortex (VLO and J,O), and the DLE and VLE, as is also the case for the HDBli.

Overall pattern of the VDB projections to the cor- fex. PHA-L injections into the caudolateral part of the VDR (VDRc) produced a paltern of labeling comparable t o that aft,er HDRrri injections, which is specifically aimed at the occipital, medial mesocortical, olfactory, lateral entorhirial, and dorsal subicular and hippocampal areas (Fig. 4). In contrast, the rostromedial parts of the VDB (VDBr) innervate the entire dorsoventral extent of t.he subicular and hippocampal areas and the entire entorhinal cortex (EC). whereas oll'actory and occipital cortices are almost, devoid of afferents from the VDBr. Furthermore, tracer delivery i n more rostromedial aspects of the VDH revealed a decreasing amount of eiferents to the medial mesocortical areas, whereas the main bulk of innervation also shifts gradually from ankrior cingulate to posterior retrosplenial cortex.

Overall pattern of the MS projections to the cortex. The pattern of terminal labeling after PHA-L injections in the MS resembles that after injections in the VDBr. All injections in the M S gave rise t,o terminal labeling mainly in the entorhinal, suhicular, and hippocampal regions, whereas some moderate projections were present in the medial mesocortical regions (Figs. 5, 6 ) . Within the subicular and hippocampal regions, regional variations in t,he innervation pattern became apparent, when comparing the medial &'IS (hs1Sm) injections (Fig. 6) with the lateral MS (MSl) cases (Fig. 5 ) . The MSl sends projections mainly to the ventral aspects of the subicular cortex (PrS, Pas . and Sub) and the hippocampus, whereas injections including t,he MSm resulted in terminal labeling in both dorsal and ventral parts of the subicular cortex and hippocampus. Such a medial-t,o-lat,eral topographic arrangement of the MS has also become apparent with respect to the projections t.o the entorhinal area, since the MSl appears to project preferentially to the ME and the VME (Fig. W,G), whereas the MSm mainly supplies the DLE (Fig. CiF,G). Only minor variations i n innervation patterns were found when comparing the dorsally and ventrally located injection sites in the MS.

Course and laminar arrangement of the MS/DB efferents to the cortex

'rhe cortical innervation patterns originating from the MS/DR complex is described on four representative PHA-L experiments. Case 5 illustrates the predominantly olfactory projections of the HDBli (Fig. 3). In case 163. efferents are described that originate from the HDBm/VDBc to illustrate the innervation patterns of the medial mesolimbic and occipital cortices and the olfactory regions (Fig. 4). Cases 195 and 118 represent the MS1 and MSm groups of injections, respectively. to illustrate the pathways and patterns of the projections to the hippocampal, suhicular, and entorhinal areas (Figs. 6 and 6).

HDBli eflerenb. All injections in lhe intermediate and lateral parts of the HDB resulted in dense anterograde labeling of project,ions in the olfactory regions, particularly in the OR? AON, and the olfactory cortex (Pir). Further- more, tracer deliveries in the HDBli provided labeled efferents in a number of mesocorl.ica1 areas as well. Experiment 5 is described to illustrate the pattern of the I-IPBli projections to the cortex. From the injection site (Fig. 3D), the mass of labeled axons radiates in anterior, ventral, lateral, and posterior directions. The Pir as well as the olfactory tubercle and the cortical amygdaloid nuclei are innervated hv t,hese axons over t,heir full antero-posterior extent (Fig. 3C-F). Axonal branches occupied with relatively large terminal bout,ons are particularly dense in the outer molecular layer I of the Pir. Thin varicose fibers are present, although less numerous in the deeper cell layers.

,4t anterior levels, the axons enter the olfactory peduncle arid terminate extensively in the AON and OB. Termina- tions are most numerous in the outer molecular layer of the AON deep the lateral ollactory tract and in the granular, mitral cell, arid periglomerular layers of the OB. A group of axoils travels in a medial direction and reaches the ventral aspects of the taenia tecta and overlying infralinibic cortex (Fig. 3C). More lateral. the orbital and anterior insular cortices receive labeled projections. A stratification of this prefrontal innervation is not obvious, although there is a slight preference for the oukr layers TIT-I. More posteriorly, the pirii'orm projections continue in the lateral entorhinal region (Fig. 3F H). The terminations are predominantly present i n the outer layer I of the DLE and VLE.

The PHA-L injection in case 163 was centered in the caudalmost part of the angular wing of the VDB, partly overlapping the HDBm (Fig. 4D). Sev- eral labeled fiber bundles can be traced from the inject,ion site. A massive group of labeled axons runs in an anterior direction subsequently coursing through the DB and the MS. In the MS, these bundles separate in two directions. One bundle of labeled fibers enters t,he dorsal fornix and fimhria, and innervates the dorsal part of the hippocampal and suhicular areas (Fig. 4E-HI. The other bundle reaches the genu of the corpus callosum from where labeled varicose fibers run medially into the TT: MO, IT,, and PL, and ven- trolat,erally into the VLO and LO, turning radially or obliquely to reach the superficial layers, while branching frequently (Fig. 4R,C). In addition, a fascicle of fibers turns caudally to join the supracallosal stria, from where they innervate the whole antero-posterior extent o f the AC and RS (Fig. 4C-H). Some labeled fibers laterally penetrate the adjacent PCm and hil.

The laminar distribution pattern of the projections in the anl.erior cingulatc areas (MO, IL, PL, and AC) is character-

HDBm/KDBc eflerents.

c3

I- N


Fig. 6. Chartings of a series of frontal sections through the rat brain (A-G) illustrating the distribution of PHA-1. labeled projections to the cortex and hippocampus following a tracer injection in the medial part

of the MS (C). Only one hemisphere is fully presented (sections E-G): although the projectinns are richly present, hut similarly distributed in bnth hcrnispheres (cf. D).


Fig. 7. Camera lucida drawing that illustrates the pattern of lamination and the predominant orientation of the PHA-L immunoreactive fibers in the retrosplenial cortex after tracer delivery in the HDBm (insert top left). The stippled area in the section shown a t the bottom left indicateb the location of the field drawn. Scaie har = 100 gm.

ized by a minor preference for the deep part of layer V and t.he superficial layer I, where the fibers are oriented predominantly in a tangential direction. Layers 11-IV are occupied mainly wit,h radially oriented fibers. This pattern of stratification is most prominent in the IL and PL. The projections in the RSg show the most pronounced stratified pattern (Figs. 4E,F, 7; see also Fig. 9c). Here, the plexus of labeled varicose fibers is generally densest in the deep part of layer V, in a small zone just below the densely packed cell layer II/III (referred to as layer IV by Eckenstein et al., '88) and occasionally also in the superficial layer I . Both in layers I and IV the fibers course tangentially, whereas in other layers they are predominantly oriented radially (Fig. 7 ) . The RSa shows a similar lamination pattern, but the more loosely packed upper cell layer II/III is more frequently invaded by branching varicose fibers.

At more posterior levels, a considerable part of the labeled fibers t,urns laterally to reach the OC (Fig. 4F-H). Here, the labeled project ions are clearly laminated. Pro- fusely hranching varicose fibers are observed predominantly in layers V, 11, and I, and in the primary visual areas (Zilles, '85). also including layer 1V (Fig. 8, 10a.b). Layers VI and 111 are relatively poor in terminal labeling. Most fibers are radially oriented and traverse the cortical layers up to the pial surface while obliquely giving off numerous varicose collaterals. These collaterals are oriented predominantly radially and obliquely in layers VI-!I and t,angentially in layers V and I (Fig. 10a). The distribution of the labeled pro-

ject,ions also display a columnar pattern, since individual labeled fibers and their branches form columnar clusters tangentially extending over approximately 200 p m . In layers V and I individual tangential varicose fibers extend up to :NO p n i n the plane of the 40-pm-thick sections. In layer IV of the primary visual area, however, the projections display marked clusters of branches extending over 200 km (Fig. 8); indicating a more strict columnar organization here.

Dorsal subicular and hippocampal areas arc richly supplied with labeled projections via the fimbria-fornix system. Labeled fibers that emanate from the fornix run into the deep layers of the dorsal PrS. Profusely branching fibers are most common in the superficial layers 111-1 of the PrS (Figs. 4H, 9e), whereas layers 1V- V1 contain numerous fibers running parallel to the cort.ical laminae. In the subiculum, the terminal labeling is distributed throughout all layers hut is most prominent in the pyramidal cell layer. The terminal labeling in the subiculiim continues in the adjacent CA1 of the hippocampus where a mare clear lamination is present (see Nvakas e t al., '87 for further details).

From the injection sites in the HDBm/VDRc, a second major group of laheled fibers can he traced in an anterior direction entering the AON and OB via the olfactory peduncle, ahundantly supplying these nuclei with terminations in a similar fashion as the HUBli projections do (Fig. 4A,B). An important difference is that, the HDBm/VDBc efferents prrferent ially terminate in the external plexiform layer and i n t,he glomeruli of the OB (cf. Figs. 3A, 4A). A third fascicle


Fig. 8. Camera lucida drawing illustrating the distribution of the PHA-L immunoreactive fihers in the primary visual cortex after tracer delivery in the HDBm (insert top left). The fibers are generally very thin with closely spaced varicosities. Note that intensive branching appears in layer IV (see also Pig. lob). The stippled area in the section shown at lhe bottom left side indicates the location of the field drawn. Scale bar =

100 urn.

of' fibcrs courses laterally and caudally and for a major part joins the medial forebrain bundle (Fig. 4E). In addition, numerous fibers run into the Fir, providing the dcep layer 111 and molecular layer I with terminal labeling, whereas the granular layer I1 is scarcely innervated. More caudally, fibers reach the cortical amygdaloid nuclei and the entorhirial cortex (Fig. 4F-H). Here, the HDBm/VDBc efferents

are mainly confined to the DLE and VLE. The terminal labeling is predominantly present in layers IV, 111, and I of the DLE. The HDHm/VL)Bc efferents are not exclusively confined tu the ipsilakral cortex and hipporampus. A few labeled libers have also been observed in the contralateral medial mesocortical, entorhinal, and subicular areas and in t,he hippocampus (Fig. 4D-F).

In cases 118 and 195, PHA-L was injected in the medial (Fig. SC) and lateral (Fig. 5C,) part of the MS, respectively. From both injection sites fascicles of labeled fibers mainly run in two opposite directions. Dor- sally running fibers merge with the dorsal fornix and fimbria to reach t,he subicular cortex and hippocampus. In addition, a limited number of axons run to the genu of the corpus callosum and innervate the II,, PL, AC, and RS in a similar fashion as the HDBm/VDB efferents do. Ventrally running MS elrerents traverse the VDB and HDBni and join the medial forebrain hundle t,o reach various subcortical forebrain and midbrain areas. A few of these labeled fibers, however, can be traced in a more lateral direction, through the deepest half of t.he lateral Pir or the amygdaloid body, to reach the entorhinal area.

The majority of the dorsally oriented labeled axons originating from t.he lateral half of the MS course through the lateral part of the fimbria supplying the ventral or temporal half of the hippocampal formation (Fig. 5E,F). Axons emanating from medial aspects of the MS travel in the more medial part, of the fimbria and the fornix to terminate i n both dorsal and ventral parts of the subicular cortex and hippocampus (Fig. 6D-G). The terminal labeling in the dentate gyrus and hippocampus proper is subject to a clear laminar distribution pattern (cf. Nyakas et al., '87). This stratification is less obvious in the Sub, although the terminal laheling is most numerous in the pyramidal cell layer. The PrS and Pas are heavily labeled throughout all layers, but densest in layers 1-111 (Fig. 9d). Radially oriented labeled fibers emanate from the angular bundle and traverse the PsS arid P r S in a predominantly radial direction while branching profusely (Fig. 32).

The main bulk of projections from the MS1 t,o the entorhinal cortex are aimed a t ME and VME, whereas substantially less fibers terminate in the DLE and VLE (Fig. 5F,G). In contrast, the entorhinal projections from the MSrn are mainly present in the DLE (Fig. 6F,G). In the ME and VME a minor stratification of the projections is present with a preference for layers IT, IV, and the upper part of V (Pigs. 5G, 9f) . In the ankrior part of the DLE, the MS efferents terminate in layers I11 and IT (Figs. 5F, 6F), but more posteriorly the labeled fibers are more evenly distributed throughout layers 1 -1V (Fig. SG).

T w o septo-entorhinal cortical pathways can be distinguished. A major pathway, which is strongly labeled after tracer injections in the MSI, runs through the limbria and the angular bundle and is aimed a t the medial parts of the eritorhinai cortex (Fig. SF,(;). A second pathway of labeled axons runs via the dorsal fornix up to the level of the forceps minor of the corpus callosum and subsequently courses in a ventrolateral direction through the deep neocortical white matter to reach the caudal DLE (Fig. 6G). Labeled axons running in the supracallosal stria have been observed up to the posterior edge of the crossing of the corpus callosum, from where they intermingle with fibers in the dorsal fornix. Therefore, these supracallosal fibers may contribute to the innervation of the entorhinal cortex and the posterior and ventral parts of t,he hippocampal formation. The second

MS efferents.


septo-entorhinal pathway through the dorsal fornix and supracallosal stria becomes strongly labeled after tracer injection in the medial par t of the MS. Axon morphology of the MS/DB efferents to

the cortex The PHA-L tracing method yields several structural fea-

tures concerning the morphology of the labeled fibers and varicosities in the cerebral cortex. There is ample evidence that the PHA-L labeled varicosities can reliably be considered as presynaptic swellings in apposition with postsynap- tic elements of t h e target neuron (Wouterlood and Groene- wegen, '85; Luiten e t al., '88). T h e great majority of the presynaptic swellings are present as the en passant type risually irregularly spaced along the fiber trunk. Also, terminal boutons positioned along or a t t h e end of short stalks originating from a parent fiber have commonly been observed throughout the cortical regions.

Another structural hallmark of the PHA-L labeled projections is the variety in the morphology of the fibers and varicosities. Such variety is richly present in the allocortical regions (Fig. IOd), but not in the neocortical regions. In the occipital cortex. as in the other neocortical regions receiving MBN-projections (Luiten e t al., '87) , the PHA-I, labeled fibers generally are very thin, bearing numerous en passant fusiform varicosities (Fig. 1Oa,b). Such fibers are also commonly present in the mesolimbic and allocortical regions. In the allocortical areas (OB, AON, Pir, EC, P a s , PrS, and Sub) , however, anot,her type of PHA-L labeled fibers is predominant. These fibers generally are more coarse and are characterized by t h e common appearance of highly branched, short terminal axon segments with large varicosities (Fig. 1Oc). Occasionally, thin axon collaterals have been observed to branch from a thicker axon (Fig. 10c,e). PHA-T, labeled fibers of both morphological classes occur intermingled in the entorhinal and subicular areas (Fig. IOd), yield- ing a midtiform axon morphology. This is in contrast with a uniform presence of innervation in the neocortical areas such as the occipital lobe (Fig. 10a,b). A similar diversity in morphology of the MS/DB efferents has been observed in the DG and hippocampus proper (Nyakas e t al., '87; Freund arid Antal, '88).

With respect to the cellular target structures of the PHA- L laheled presynaptic boutons, the far greatest majority of the terminal structures d o not closely appose t.he cell bodies, thus suggesting preponderant synaptic contacts with distal dendritic profiles. Terminal boutons embracing neuronal cell bodies have only rarely been observed. Besides these interneural contacts, labeled terminations occasionally oc- curred in close proximity to the cerebral microvasculature in all cortical target, fields (Fig. lOe>f).

Comparison of the MS/DB efferents with the AChE- and CMT-positive fiber patterns in the

cortex

T h e results of t h e AChE- and ChAT fiber staining are mentioned only briefly in this report for comparison with the PHA-T, tracing experiments. T h e ChAT-immunoreactivity patterns in the cerebral cortex and hippocampus as revealed in our experiments generally agree with recently reported demonstration of this cholinergic marker (Houser et at., '85; Parnavelas et al., '86; Brady and Vaughn, '88; Eck- enst,ein et, al.. '88; T,ysakowski e t al., '89). In our observations the pattern of AChE positive fibers appeared to be generally

identical to the distribution of cholinergic innervation profiles demonstrated wit,h ChAT-immunoreactivity in the cor- t.ical mantle, which does not account for the cortical cell bodies stained for ChAT or AChE (Fig. 9a.b). Recently, Lysakowski e t al. ('89) carried out a comparative survey study of ChAT- and AChE-positive profiles in the ra t cortex. They described identical distribution patterns of ChAT and AChE in most areas, with some minor differences in the barrel-field somatosensory, cingulate, and relrosplenial cortices. A remarkable difference between AChF, and ChAT staining was observed by these authors in t,he retrosplenial areas. In contrast to their findings, our material revealed a near identical distribution of AChE- and ChAT-stained fibers in the RSg (Fig 9a.h) also designated as area 29b by Vogt and Peters ('81). In both our AChE- and ChAT- stained sect,ions, a narrow but dense plexus of fibers is present just. deep to the dense cell layer ll/III. Furthermore, layer I is heavily innervated with predominantly tangential- oriented fibers, whereas layer V contains a moderately dense contingent, of mainly radially and obliquely running fibers. Together with the upper par t of layer V, layer II/III is least. densely innervated with fibers. T,ayer VI contains an addit,ional group of fibers, which emerge from the cingulum. T h e RSa (area 29c of Vogt and Peters, '81) displays a comparable pattern, but, the dense layer IV plexus is less obvious and the more loosely packed layer II/III is moderately dense innervated like layer V.

In most other cortical areas we have observed AChE- and ChAT-stained fiber patterns generally similar to t.he de- scriptions of Lysakowski e t al. ('89). However, our observations concerning the cholinergic innervation of the primary visual cortex (area 17 ) agree well with t,he description of Eckenstein et, al. ('88, cf. Fig. 3 of their paper), who stress the presence of a dense ChA1'-positive fiber plexus over layer 1V.

A comparison of the PHA-L labeled cortical afferents and cholinergic cortical innervation as visualized by AChE- or ChAT-staining revealed similar patterns of fiber distribution in most cortical areas, but, also showed some remarkable differences. T h e minor degree of lamination of the MS/DB projections t o the anterior mesocortical areas (ACd, ACV, VI,O, LO, and AId,v) agree well with relative small variations in AChE/ChAT-fiber density in these areas. According to the AChEiChAT-fiber distribution, we have observed a slight pref'erence of PHA-L labeled innervation for the outer layer I of the IT, and PL. Concordantly, a clear stratification of the MS/DB rlferents occurs in the retrosplenial areas, where the labeled fibers are part,iciilarly numerous in a narrow zone deep t o layer II/III (also demarcated as layer IV) and in the deep par t of layer V (Fig. 7,Sc). In contrast, to the AChE/ChA'I'-fiber distribution, layer I is not as densely innervated. Also in line with the AChE/ChAT-fiber distribution. a clear stratification of the PHA-L laheled projections occurs in the occipital cortex, where terminal labeling is preferentially present in layer V, the upper par t of layer 111. and in layers I1 and I. T h e primary visual area receives an additional labeled input in layer IV and in t,hat, sense dif- fers from the lamination of the MBN efferents in primary sensory fields of other modalities (Luiten e t al., '85, '87). MS/I)B projections in the suhiciilar and entorhinal areas show some degree of lamination comparable to that of t.he AChE/ChA'Y-stained fibers. As is particularly the case in the EC, these patterns are far less pronounced. PHA-L labeled projections t o the AON and Pir are most richly present i n layer I, which is in contrast to the distribution of cho-


Fig. 9. Photomicrographs illustrating the distribution of cholinergic fihers in the retrosplenial cortex stained for AChE (a) or ChAT (b), and the PHA-I. labeled axuns and terminals in the ret,rosplenial (cj, subicular (d,e) and entorhinal cortex ( f ) . (a-c) The laminar preference of the terminal laheling in layer I\' of the RSg coincides with that of the cholinergic fibers, either stained for AChE or ChAT (see also Fig. 7). (d)

Labeled projections in t,he PaS after injection of the tracer in the MS1. (e ) PHA-L immunoreactive axons in the dorsal PrS after tracer injection in the HDBm. (f) Massive terminal labeling in the VME after a large PHA-L injection in the MS, which is particularly present in layer 11. Scale bars in all panels = 100 p m .


Figure 10


linrrgic fibers with highest density in and around layer 11. T h e distrihution of PHA-L labeled projections from the HDBli and AChE/ChA'l-positive fibers in the OB was shown to be similar.

DISCUSSION In this study t h e projections from the hlS/DR complex t o

the cortical mantle have been investigated in detail with the anterograde PHA-L tracing method. T h e present experiments showed rich innervation of the allo- and mesocortical areas and the occipital neocortex. I t has also been demonstrated that remarkable differences in regional projection pat terns throughout the cortex occur dependent on the location of the PHA-L injections within the MS/DR complex, which reflects a topographic organization of t h e MS/L)B- cortical connectivity. In addition to this characteristic areal distribution, the PHA-L labeled efferents also showed distinct, laminar innervation patterns within the various c o r t - cal areas.

Topographic organization of the MS/DB complex

T h e topographic arrangement of the MS/DB neurons with respect to their connections with the cortical mantle is in register with the results of various studies using retrograde axonal tracing (Swanson and Cowan, '79; Macrides e t al., '81; McKinney e t al., '83; Mesillam e t al., '83b; Alonso and Kiihler, '84; Rye e t al., '84; Saper: '84; Woolf et a]., '84; Amaral and Kurz, '85; ZAborsky et ai., '86). T h e major characteristics of this anatomical organization can he summa- rized as follows.

First, the HDH as a whole and VDBc are the sources of projections to the olfactory structures (Figs. 3, 4, 1 lA,Bj, whereas the MS, VDB, and HDBm constitute a continuous string of nuclei tha t innervates the hippocampal region including the subicular and entorhinal cortices (Pigs. 4, 5,6, l lB ,C) . Thus a considerable intermingling of cells projecting to the olfactory and hippocampal st,ructiires occurs in the HDBm and adjacent VDBc (Figs. 4, 1lB). Individual cholinergir cells in this par t of the I)L( have recently been reported t o give rise to collateral projections t,o hot,h olfactory bulb and hippocampus (Okoyama et al., '87j.

Second, the MS/DB complex is also a major source of input to the mesocortex. The projections t o the medial mesocortex emanate from most parts of the MS/DB complex. 'I'he majority of the medial mesocortical input, however, comes from the HDRm and VDB, where a reverse anterior-posterior topography is present, such that more

Fig. 10. Photomicrographs demonstrating the morphology of the PHA-L labeled fibers in several cortical areas. (a,b) Thin varicose fibers appear in layers I (a) and TV (b) of the occipital visual cortex after tracer delivery in the HDBm. Note the predominant orientation of the fibers (see also Fig. 8). (c,d) PHA-I, labeled MS efferents in the superficial layers of the PrS; (c) illustrates the branching of a coarse axon (arrowhead) into thin axon collaterals, which in turn branch off short stalks with terminal boutons (arrows); (d) shows the appearance of morphological distinct types of fibers in the PrS. Both thin varicose (arrow) and thick coarse fibers (arrowhead) are present. (e,f) PHA-L labeled efferents in the ME. The terminal laheling appears in close proximity to the cerebral microvasculature (arrow in e) . In panel e axon collaterals diverge Crom a parent axon (arrowhead). Scale bars in all panels = 50 m.

posteriorly located cells in the VDB and HDBm project t o more anterior levels of the AC and RS (Fig. l lB,C). T h e majority of the input to t.he lateral mesocortices comes from the HDR (Fig. I IA) . T h e distrihrition of the orbital- and insular corticopetal cells resernhles the location of the neurons innervating the olfactory nuclei (Table I).

'Third. the HUB and VDR send considerable numbers of etferents to circumscribed parts of the neocortex. T h e occipital cortex, which is almost devoid of input from the M B N (Liiit,en et, al., ' 87 ) , receives its (cholinergic) aff'erents predominantly from the HDBm and VDBc (Figs. 4, 11A). Apart from the OC, the neocortical regions to receive some MS/DB input are those neighboring the mesocortical areas, i.e.. the PCm and the most medial par t of t,he M and the ventral part of the SS and Te. So it can be concluded tha t the MS/DR projcct,ions arc not cxclusively restricted to the allo- and mesocortical domains. Likewise, the projections of the MRN are not confined to the frontal. parietal, and t,emporal neocortices, but extend into the mesocortex as well (Luiten et. al., '85, '87).

A more detailed analysis of the MS/DB projections to the hippocampal formation reveals a strong topographic ar - rangement of the projections along the septo-temporal axis of the hippocampal and subicular areas. Sept,al levels of the hipporam pus receive projections primarily from the VDBc, HDBm, and the rnedial. largely noncholinergic portion of the MS (Figs. 3 : 6, 1 I R , C j , whereas the VDBr and RilSl innervate more temporal levels of the hippocampal formation (Figs. 5. 11C). In this respect our findings are in line with several retrograde tracing studies (Segal and Landis, '74; Meihach and Siegel, '77; Amaral and Kurz, '85), which showed a medial-to-lateral topographic arrangement of the septo-hippocampal projection.

The entorhiiial cortex takes an exceptional position with respect to the organization of the MS/DR-cortical pro- .jection. The rostrolateral portion of the EC is one of t,he few cortical regions t o receive efferents from essentially all suhdivisions of t.he MS/nB complex, and from the MBN (Luiten et al., '87). The present study demonstrates a topographic arrangement of the MS/DR-entorhinal projection with regard to the rnediolateral axis of the EC. T h e HDR and VDRc mainly innervate the lateral zones of the EC (Fig. 11A.B). whereas the VDBr and MS project to both medial arid lateral parts of the EC (Fig. 110. T h e MS1 preferentially projects t o the medial aspects of the EC, whereas the MSm ahuridant,ly innervates the lateral aspects of t,he EC (Figs. 5. 6, 11 Cj. This topographic relation, however, is in contrast with the findings of Alonso and Kiihler ('84). These conflicting findings may possibly be the result of leakage of the retrograde trarer to the adjacent ventral hippocampal area in Alonso and Kiihler's study, an area known to receive massive projections mainly from the lateral parts of the MS (vide supra, Segal and Landis, '74; Meibach and Siegel, '77; Amaral and Kurz, '85; Nyakas e t al., '87).

In agreement with the results of previous studies (Milner e t al., '83; Milner and Amaral, '84). the present experiments demonstrate three major pathways of the MS/DB projection to the cortical; hippocampal. and olfactory regions. T h e septa1 fibers run predominantly via the fimbria-fornix system t o reach all hippocampal structures (Pig. 11C). In addition, more posterior parts of the hippocampal region are innervated hy septa1 axoris running via the supracallosal buridle (Fig. l lB ,C) . This latter bundle comprises the bulk of axons projecting to the niedial meso- and the occipital neocortex. Axons emanating from the HDR and VDB run


A

Fig. 11. Schematic representation of the major pathways and projections oC the MS/DB complex to the olfactory, hippocampal and cortical areas, as revealed by the anterograde transport of PHA-L in this study,

which are topographically subdivided into the HDBli (A), HDBm, VDBc (B) , VDHr. MS1 and MSm (C).


C

into the olfactory peduncle to reach the olfactory bulb. Resides, VDB and HDR fibers course through a ventral/lat- era1 pathway to innervate the piriform cortex, the insular and orbital areas anteriorly, and the entorhinal and ventral subicular areas posteriorly (Fig. 1 IA,B).

The topographic principles of the MS/DB projection as described here match the general organization of the MBN (Luiten et al., ’87). The most anterior subdivision of the MRN (MBNa) preferentially project to the dorsomedial aspects ofthe cortical mantle, including ACd, PL, and IL. In this respect, the HDBm and VDB may thus be considered as an anteromedial continuation of the MBNa, since these parts of the MS/DB complex innervate the medial mesolimbic cortex most heavily. Similarly, the HDBIi is also continuous with the adjacent ventral aspects of the MBN with respcct to the innervation of the olfactory and insular cortical areas.

Cholinergic versus noncholinergic MS/DB projections

The PHA-L labeled efferents generallr were present as thin varicose profusely branching fibers throughout all layers of the meso- and neocortical target areas. The morphology, laminar orientation. and distribution of the labeled fibers in the cortical laminae in most cases closely match the cortical cholinergic innervation as displayed by AChE-his- tochemistry or by ChAT-immunoreactivity as revealed by the present experiments (Fig. 9a,b) and previous studies (Houser et al., ’85; Parnavelas e t al., ’86; Brady and Vaughn, ’88: Eckenstein et al., ’88; 1,ysakowski et al., ’89). These similarities support the evidence, that the majority of the meso- and neocortical projections of the MS/DB complex is

cholinergic as was demonstrated by retrograde tracing in combination with ChAT-immunohistochemistry (Rye et al., ‘84; Woolf et al., ’84). However, some striking differences between the distribution of cholinergic fibers and PHA-L filled projections also have appeared in several allocortical, notably the olfactory, hippocampal (cf. Nyakas et al., ’87 and Matthews et al., ’87) and the entorhinal areas. I t is well documented that both cholinergic and noncholinergic neurons in the MS/DB complex project to the olfactory and (para-) limbic rnesocortical areas (Rye et al., ’84; Woolf et al., ’84; ZBborsky et al., ’86) and the subicular and entorhinal cortices (Alonso and Kiihler, ’84; Amaral and Kurz, ’85). Glutanric acid decarhoxylase- and GARA-immunoreactive neurons have also been found to project to the allocortical regions as the OB (Zaborsky et al., ’86). hippocampus, and entorhinal cortex (Kiihler et al., ’84; Freund and Antal, ’88).

Cortical projections originating from these noncholinergic. partly GABAergic cells of the MS/DB complex may contribute to the ohserved discrepancies between the anterogradely PHA-L labeled arid ChAT-immunoreactive innervation. The concomitant presence of cholinergic and noncholinergic projections may also be responsible for the morphological heterogeneity of the anterogradely laheled fibers i n the allocortical regions (Fig. IOc-e). The neocortical areas are supplied with mainly thin axons with generally small varicosities (Fig. 10a,b) that are comparable to ChAT- positive axons. The distribution of this type of axons (designated “type 2” by Nyakas et al., ’87) more closely matches the cholinergic innervation in the olfactory, hippocampal, and entorhinal areas. For these arguments it is suggested that the fine “type 2” axonal network is cholinergic, whereas the thicker “type 1” axons characterized by the presence of


larger boutons are the noncholinergic counterpart of the MS/DR-cortical projection. Freund and Antal ('88) demon- strat,ed directly the presence of GABA in the "type 1" septo- hippocampal projections containing large houtons (cf. Nya- kas et al., '87) , which formed multiple basketlike contacts around cell bodies and proximal dendrites of the CA3 GABAergic interneurons.

I t is hazardous, however, to differentiate the PHA-L labeled projections to the allocortex into two separate cholinergic and noncholinergic axonal systems, based only on their morphological characteristics. The neurochemical identity of the PHA-L labeled fibers therefore merits further st,urly, in which a dual labeling of PHA-L and ChAT in the axonal processes may provide additional information (Gerfen and Sawchenko, '85; Woolf et al., '86).

Discrepancies between laminar patterns of the ChAT- positive and the PHA-I, labeled fibers arising from the basal forebrain may also be explained by the presence of an intrinsic cholinergic circuit in the rodent cortex and hippocampus (Houser et al., '83, '85; Levey et al., '84; Wainer et al.. '85; Frotscher et al., '86; Parnavelas et al., '86; Matthews et al., '87; Blaker e t al., '88; Eckenstein et al., '88). However, the cortical and hippocampal small ChAT-positive, generally bipolar interneurons do not contribute to the characteristic laminar patterns of the ChAT-immunoreactive fihers in the neocortical areas (Eckenstein e t al., '88) or in the hippocampus (Matthews et al., '87). This is due to the fact that the axons of the ChAT-positive interneurons were reported t,o stain extremely faint and difficult t,o detect in the cortex (Eckenstein et al., '88) or hippocampus (Matthews et, al., '87). I t has also been mentioned that caution must be taken using AChE as a marker of extrinsic cholinergic neurotrans- mission in the cortex and hippocampus, since this enzyme has been detected in the cortical and hippocampal neurons as well. Complete subcortical deafferentation of the cortex or hippocampus was, however, reported to result in an elimi- nation of virtually all AChE-containing fibers (Bear e l al., '85; Matthews et al., '87; Blaker et al., '88). These ohserva- tions are in line with the fact that the AChE-positive neurons in the neocortex do not give rise to AChE-stained axons (Rear et al., '85). Moreover, AChE was reported to be essentially absent within the hippocampal (Matthews et al., '87) and cortical ChAT-positive interneurons (Levey e t al., '84; Blaker e t al., '88). Therefore, it is likely that t,he intensely AC,hE-stained fibers in the cortex and hippocampus, as well as the strongly ChAT-immunolabeled axons and terminals, are exclusively of .whcortical origin and most likely represent the same cholinergic cortico- and hippocampopetal system. This cholinergic projection essentially arises from the basal forebrain, although a minor projection from the cholinergic cells in the pontomesencephalon is present (Mesu- lam et al., '83b; Vincent et al., '83).

Nomenclature of the MS/DB complex The present study shows a detailed survey of the topo-

graphic principles of the RIISIDB complex with respect to t.heir cortical, hippocampal, and olfactory projections, which may add to define an appropriate parcellation of the MS/ DB complex in the rat. The topographic parcellation of the MS/DB complex with respect to the cortical projection yields three major suhdivisions (HDBIi, HDBm/VDBc, and VDBr/MS), each giving rise to a particular distribution of cortical innervation. An interesting aspect, concerning the subdivision of the HDB into a lateral-intermediate (HDBli) and a medial portion (HDBm), is the presence of remark-

ahle differences between the distribution of cortical efferents from these compartments. In line with these differences in efferent connections, a number of cell morphological and neurochemical differences between the HDBli and the HDBm have been reported (Swanson, '76; Zaborsky e t al., '86). The nomenclature used by Swanson ('76), in which the HDRli has been indicated as the "magnocellular preoptk region" (MgPO) separated from the more medially located "horizontal limb o f the diagonal band" (as defined HDBm and VDB in our study), therefore seems reasonable, hut is still confusing for two reasons. First, both HDRm and HDRli are strongly connected with the olfactory regions, and as such can he considered collectively as one nucleus and an anatomical entity. Second, the use of the term MgPO suggests a stronger relationship of the HDBli with the lateral and medial preoptic regions, which show connect.ivity patterns substantially dilferent from the HDBli in sending no efferents to the cortical mantle (own observations, Swan- son, '79).

Mesulam et al. ('83b) proposed a Ch-nomenclature for the cholinergic cell groups in the rat basal forebrain, in which the MS, VDB, HDB, and the MBN as defined in this study were labeled Chl to Ch4, respectively, homologous to such a subdivision in the monkey basal forebrain (Mesulam e t al., '83a). This parcellation is based on the consideration that the HDR contains a vast majority of the olfactory hulbope-

ion neurons, whereas the VDR and MS are the only sources of the hippocampal projections within the basal forebrain. From the present results, we conclude that this Ch-classification for the rat basal forebrain is not sufficient to match the complex topography in every respect. This is hest illustrated by the cortical projection pattern of the VDHc (pars angularis of t.he DB) and the HDBm. The HT)Rm and VDRc contain a mosaic of cells aiming a t the olfactory bulb, the hippocampus, and the medial mesolimbic and occipital cortices. The projection patterns of these latter two basal forebrain sectors show that the rat basal forebrain corticopetal system might not he differentiat,ed to the same degree as in t,he primate. The rat homologue of the monkey Ch4am-characterized by its projections to the visual and medial aspects of the cortex (Mesulam et al., '83a)-is highly intermingled with the bulbopetal (Ch3) and hippocampal (Ch2) neurons in the HDBm/VDBc, which thus cannot easily be labeled as a single Ch-group.

ACKNOWLEDGMENTS The authors thank Dr. F. Wouterlood for his contribution

to the experimental material, and Dr. Bruce Wainer for his generous g i f t of th r monoclonal anti-ChAT antibody. Dr. T. Matsuyama is acknowledged for his contribution to the ChAT inimunocytochemistry.

LITERATURE CITED Alonso, A., and C. Kohler (1984) 4 study of the reciprocal connections

between the septum and the entorhinal area using anterograde and retrograde axonal transport methods in t.he rat brain. J. Comp. Neurol. 225:327-343.

Amaral, D.G., and ,J. Kurz (1985) An analysis of the origins of the cholinergic and noncholinergic septa1 projections to the hippocampal formation of the rat. J. Cnmp. Neural. 240:37-59.

Arendt, T., V. Bigl, A. Tennsted, and A. Arendt (1985) Neuronal loss in different parts of the nucleus basalis is related to neuritic plaque formation i n cortical target areas in Alzheimer's disease. Neuroscience 14:l- 14.

Rarr.iis, R.T.. R.L. Dean 111, B. Beer, and A S . Lippa (1982) The cholinergic hrpothesis of geriatric memory dysfunction. Science ZI7:40€L417.


Hear. M.F., K.M. Carries, and F.F. Ehner (1985) An investigation of &liner. gic CjrCUitY in Cat striate cortex using acetylcholinesterase histoehemistry. J . Comp. Neurol. 234:411-430.

Hid, v., N. Woolf. and L.L. Butcher (1982) Cholinergic projections from the basal forebrain to frontal, parietal, temporal, occipital and cingulatt. cortices. a combined fluorescent tracer and acetylcholinesterase analysis, Brain Res. Bull. 85'27- 749.

B l a h , S.N., D.M. Armstrong, and F.H. Gage (1 988) Cholinergic neurons within the rat hippocampus: Response to fimhria-fornix transection, J. Comp. Neural. 272.127-138.

Rrady, D.R., and J.E. Vaughn (1988) A comparison of the localization ofcho. line acetyltransferase and glutamate decarboxylase immunoreactivity in rat cerebral cortex. Neuroscience 24:1009-1026.

Butcher, L.L., K. TalboL, and L. Bilezikjian (1975) AcetylcholinesLerase neurons in dopamine-containing regions of the hrain. J . Neural Trans. 37:127 153.

Code, cJ.T., D. Price, and M.R. DeLong (1983) Alzheimer's disease: A &or. der of cortical cholinergic innervation. Science 2191184-1190.

Eckenstein, F.. and M.V. Sofroniew (1983) Identification of central cholinergic neurons containing both choline acetyltransferase and acetylcholinesterase and of central neuron8 containing only acetylcholinesterase. J. Nrurosci. Jt2286-2291.

Rrkenstein, F.P., R.W. Baughman, and J. Quinn (1988) An anatomical study of cholinergic innervation in rat cerebral cortex. Neuroscience 25:457- 47.2.

Flicker, C.. R.L. Dean, D.L. Watkins, S.K. Fisher, and R.T. Bartus (1983) Behavioral and neurochemical effects following neurotoxic lesions of a major cholinergic input to the cerehral cortex in the rat. Pharmacol. Bio- chem. Behav. 18:973-981.

Freund, T.F., and M. Antal (1988) GABA-containing neurons in the septum control inhihitory interneurons in the hippocampus. Nature 336:170- 173.

Frotsrher, M., and C. Lkranth (1985) Cholinergic innervation of the rat hippocampus as revealed by choline acetyltransferase immunocytochemis- try: a combined light and electron microscopic study. J. Comp. Neural. 239:237-246.

Friitscher, M., M. Schlander, and C. LBr6nth (1986) Cholinergic neurons in the hippocampus. A cvmbined light and electron microscopic immunocytochemical study in the rat. Cell Tissue Res. 246293-301.

Gerfen, C.R.. and P.E. Sawchenko (1984) An anterograde neuroanatomical tracing method that shows the detailed morphology of neurons, their axons and their terminals: immunohistochemical localization of an axonally transported plant lectin. Phaseolus oulgaris leucoagglutinin (PHA- L). Brain Res. 290:219-238.

Gerfen. C.R.. and P.E. Sawchenko (1985) A method for anterograde axonal tracing of chemically specified circuits in the central nervous system: combined Phaseolus vulgaris leucoagglutinin (PHA-L) tract tracing and immunohistochemistry. Brain Res. .?4.'?144-150.

Hedreen, J.C., S.J. Bacon, and D.L. Price (1985) A modified histochemical technique to visualize acetylcholinesterasecontaining axons. .J. his to^ chem. Cytochem. 33t134-140.

Hellendall, R.P., D.A. Godfrey, C.D. Ross, D.M. Armstrong, and J.1,. Price (1986) The distribution of choline acetyltransferase in the rat amygdaloid complex and adjacent cortical areas, as determined by quantitative microassay and immunohistirchemistry. J. Comp. Keurol. 249.488-498.

Henderson. Z. (1981) A projertion from acetylcholinesterase containing neurons in the diagonal hand to the occipital cortex of the rat. Neuroscience 6r1081-1088.

Houser, C.R., G.D. Crawford, P.M. Salvaterra, and .J.E. Vaughll (1985) Immunocytochemiral localization of choline acetyltransferase in rat cerebral cortex: A study of cholinergic neurons and synapses. J . Comp. Neu- rol. 234r17-34.

Houser, C.R., G.D. Crawford, R.P. Barber, P.M. Salvaterra, and d.E. Vaughn (1983) Organization and morphological characteristics of cholinergic neurons: an immunocytochemical study with a monoclonal antibody to choline acetyltransferase. Brain Res. 266:97-119.

Johnston, M.V., A.C. Young, and J.T. Coyle (1981) Laminar distribution of cholinergic markers in neocortex: effects of lesions. J. Neurosci. Res. 6:597-607.

Kohler, C., V. Chan-Palay, and d.Y. Wu (1984) Septa1 neurons containing glutamic acid decarboxylase irnmunoreactivity project to the hippocampal region in the rat brain. Anat. Embryol. l69t41-44.

Krettek. J.E., and J.L. Price (1977a) The cortical projections of the me[& dorsal nucleus and adjacent thalamic nuclei in the rat. J. Comp. Neural. 171rlS7-192.

Krettek J.E., and J.L. Price (1977b) Projections from the amygdaloid plex and adjacent olfactory structures to the entorhinal cortex and to the subiculuin in the rat and cat. tJ. Comp. Neurol. 172:687-722.

I,amour, IT., P. Dutar, and A. Johert (1984) Cortical projections of the nucleus ofthe diagonal band of Broca and of the substantia innorninata in the rat: An anatomical study using the anterograde transport [Bf a conjugate of wheat germ agglutinin and horseradish peroxidase. Neuroscience 12.395- 408.

LeveY, A.I., B.H. Wainer, E.J. Mufson, and M.-M. Mesulam (1983a) ~ ~ 1 ~ ~ ~ 1 . ization of acetylcholinesterase and choline acetyltransferase in the rat cerebrum. Neuroscience 9t9-22.

Levey, A.I., D.M. Armstrong, S.F. Atweh, R.D. Terry, and B.H. Wainer (1983b3 Monoclonal antibodies to choline acetyltransferasc: production specificity and immunohistochemistry. J. Neurosci. 3t1-9.

LeveY, A.I., R.H. Wajner, D.B. Rye, E.J. Mufson, and M.-M. Mesulam (1984) Choline acetyltransferase~immunoreactive neurons intrinsic to rodent cortex and distinction from acetylcholinesterase-positive neurons. Neu- roscience 13 341-363.

Lewis. P.R., and C.C.D. Shute (1967) The cholinergic limbic system: projections to hippocampal formation, medial cortex, nuclei of the ascending cholinergic reticular system, and the suhfornical organ and supra-uptic crest. Brain 9Ot521-540.

Luitm, P.G.M., H.P.A. Gaykema, J. Traber, andD.G. Spencer Jr. (1987) Cor- tical proiection patterns of magnocellular basal nucleus subdivisions as revealed by anterogradely transported Phaseolus uulgaris leucoagglutinin. Brain Res. 413:229-250.

Luiten, P.G.M., D.G. Spencer Jr., d. Traber, andR.P.A. Gaykema (1985) The pattern of cortical projections from the intermediate parts of the magnocellular nucleus hasalis in the rat demonstrated by tracing with Phaseulus vulgaris leuco-agglutinin. Neurosci. Lett. 57:137-142.

Luiten, P.ti.M., F.G. Wouterlood, T. Matsuyama, A.D. Strosberg, B. RIP walda. and R.P.A. Gaykema (1988) Immunocytochemical applications in neuroanatomy. Demonstration of connections, transmitters and recep- tors. Histochemist,ry ,9035- 9'7.

Lysakowski, A., R.H. Wainer, G. Bruce, and L.B. Hersh (1989) An atlas of the regional and laminar distrihut,ion of choline acetyltransferase immunoreactivity in rat cerebral cortex. Neuroscience 28.291-336.

I.ysakowski, A., B.H. Wainer, D.R. Rye, G. Bruce, and L.B. Iiersh (1986) Cholinergic innervation displays strikingly different laminar preferences in several cortical areas. Neurosci. 1.ett. 64:102-108.

McKinney, M., J.T. Coyle, and J.C. Hedreen (1983) Topographic analysis of the innervation of the rat neocortex and hippocampus by the hasal forebrain cholinergir system. J. Comp. Neurol. 217:103-121.

Macrides, F., R.S. Davis, W.M. Youngs, N.S. Nadi, and F.1,. Margolis (1981) Cholinergic and catecholaminergic afferents to the olfactory bulb in the hamster: A neuroanatomical, biochemical and histochemical investigation. J . Comp. Neurol. 203t495-514.

Matthews, D.A., P.M. Salvaterra, G.D. Crawford, C.R. Houser, and J.E. Vaughn (1987) An immunocytochemical study of choline acetyltransferase containing neurons and axon terminals in normal and partiallv deaf- ferented hippocampal formation. Brain Res. 402.30-43.

Meibach, R.C., and A. Siege1 (1977) Efferent connections of the septal area in the rat: an analysis utilizing retrograde and anterograde transport met,h- ods. Brain Res. 119.1-20.

Mellgren, S.I., and B. Srebro (1973) Changes in acetylcholinesterase and distribution of degenerating fibers in the hippocampal region after septal lesions in the rat. Brain Res. 52:19-36.

Mesulam, M.-M., A.D. Rosen, and E.J. Mufson (1984) Regional variations in cortical cholinergic innervation: chemoarchitectonics of acetylcholinesterase-containing fibers in the macaque hrain. Brain Res. 311245-258.

\fesulam. M-M.. E.J. Mufson, A.I. Levey, and B.H. Wainer (1983a) Cholin- ergic innervation of cortex by the hasal forebrain: Cytochemist,ry and cor- t.ical connections of the septal area, diagonal hand nuclei, nucleus basalis (substantia innominata), and hypothalamus in the rhesus monkey. J. Comp. Netmil. 214.170-197.

Mesulam, M-M., E.J. Mufson, B.H. Wainer, and A.I. Levey (1983h) Central choliriergic pathways in the rat: an overview based on an alternative nomenclat.ure Chl-Chi? Neuroscience lOr1185-1201.

MiIrier, T.A., and D.G. Amaral (1984) Evidence for a ventral septal projection to t,he hipporampal formation of the rat. Exp. Brain Res. 55r579-585.

Milner, T.A., R. Lay, and D.G. Amaral(1983) An anatomical study ofthe sep- tohippncampal projection in the rat. Dev. Brain Res. XtR48-371.

Nyakas, C., P.G.M. Luiten, D.G. Spencer, and J. Traher (1987) Detailed projection patterns of septal and diagonal band efferents to the hippocampus in the rat with emphasis on innervation of CA1 and dentate gyrus. Brain


Oknyama, S., H. Tago, P.L. McGeer, L.B. Hersh, and H. Kimura (1987) Cho- linergic divergent projections from rat hasal forebrain to the hippocampus and olfactory bulb. Neurosci. Lett. R3r77-81.

Parnavelas, J.G., W. Kelly, E. Franke, and F. Eckenstein (1986) Cholinergic neurons and fihers in the rat visual rortex. J . Neurocytol. 15:329-336.

Paxinos, G., and C. Watson (1986) The Rat Brain in Stereotaxic Coordinates. New York: Academic Press.

Reep, R. (1984) Relationship between prefrontal and limbic cortex, a comparative review. Brain Behav. Evol. 255-80.

Rye,D.B.,B.H. Wainer, M.-M. Mesulam,E.J. Mufson,andC.B. Saper (1984) Cortical projections arising from the basal forebrain: a study of cholinergic and noncholinergic components employing combined retrograde tracing and iminunohistochemical localization of choline acetyltransferase. Neuroscience 13,627 -643.

Saper, C.B. (1984) Organization of cerebral cortical afferent systems in the rat 11. Magnocellular hasal nucleus. J . Comp. Neurol. 22231 1-342.

Saprr, C.B.. D.C. German, and C.L. White 111 (1985) Neuronal pathology in Lhe nucleus basalis and associated cell groups in senile dementia of the Alzheimer’s type. Neurology 351089-1095.

Segal, M., and S. Landis (1974) Afferents to the hippocampus of the rat studied with the method of retrograde transport of horseradish peroxidase. Brain Res. 78:l-15.

Shute, C.C.D., and P.R. Lewis (1967) The ascending cholinergic reticular sys- k m : Neocortically, olfactory and subcortical projections. Brain 90:497- 620.

Spencer, Jr., D.G., and H. La1 (1983) Effects of anticholinergic drugs on learning and memory. Drug Dev. Res. 3;489-502.

Spencer, Jr., U.G., E. Horvith, P.G.M. Luiten. T. Schuurman, and J. Traber (1985) Novel approaches in the study of brain acetylcholine function: Neuropharmacology, neuroanatomy and behavior. In J. Traher and W.H. Gispen (eds): Senile Dementia of the Alaheimer Type. Berlin: Springer, pp. 325-342.

Storm-Mathisen, J . , and T.W. Blackstad (1964) Cholinesterase in the hippocampal region. AcLa Anat. 56:216-253.

Swanson, L.W. (1976) An autoradiographic study of the efferent connections of thr preoptic region in the rat. .J. Comp. Neurol. 167;227-256.

Swanson. L.W., and W.M. Cowan (1979) The connections of the septa1 region in the rat. J. Comp. Neurol. 186r621-656.

Ter Horst. G.J., H.J. Groenewegen, H. Karst, and P.G.M. Luiten (1984) Pha- s d u s uulgaris leuco-agglutinin immunohistochemistry. A comparison

between autoradiographic and lectin tracing of neuronal efferents. Brain Hes. 307379-383.

Vincent, S.R., K. Satoh, D.M. Armstrong, and H.C. Fibiger (1983) Substance P in the ascending rholinergic reticular system. Nature 306,688-691.

Vogt, B.A., and A. Peters (1981) Form and distribution of neurons in rat cingulate cortex: Areas 32, 24 and 29. J. Comp. Neurol. 195:603-6‘25.

Wainer, B.H.. A.I. Levey, E.J. Mufson, and M.-M. Mesularn (1984a) Cholin- ergic systems in mammalian brain identified with antibodies against choline acetyltransferase. Neurochem. Int. 6163-182.

Wainer, B.H., A.I. Levey, D.B. Rye, M.-M. Mesulam, and E.J. Mufson (19851 Cholinergic and nun-cholinergic septohippocarnpal pathways. Neurosci. Lett. 54:45-52.

Wainer, B.H., J.P. Bolam, T.F. Freund, Z. Henderson, S. Totterdal, and A.D. Smith (1984b) Cholinergic synapses in the rat brain: a correlated light and electron microscopic immunohistochemical study employing a monoclonal antibody against choline acetyltransferase. Brain Res. 908:69-76.

Wenk, H., V. Bigl, and U. Meyer (1980) Cholinergic projections from magnocellular nuclei of the basal forebrain to cortical areas in rats. Brain Kes. Rev. 2.295-316.

Whitehouse, P..J., D.1,. Price, R.G. Struble, A.W. Clark, J.T. Coyle, and M.R. DeLong (1983) Alzheimer’s disease and senile dementia: Loss of neurons in the basal forebrain. Science 215:1227-1239.

Woolf, N.d., F. Eckenstein, and L.L. Butcher (1984) Cholinergic systems in the rat brain: I. Projections to the limbic telencephalon. Brain Res. Bull. 131751-784.

Woolf, N.J., M.C. Hermit, and L.L. Butcher (1986) Cholinergic and noncholinergic projections from the rat basal forebrain revealed by combined choline acetyltransferase and Phaseolus uulgaris leucoagglutinin immii- nohistochemistry. Neurosci. Lelt. 66:281-286.

Wouterlood, F.G., and H.J. Croenewegen (1985) Neuroanatomical tracing by use of Phaspolus uulgaris leucoagglutinin (PHA-L): Electron microscopy of PHA-L filled neuronal somata, dendrites, axons and axon terminals. Brain Res. 926:188-192.

Zaborsky, I,., .I. Carlsen. H.R. Brashear, and L. Heimer (1986) Cholinergic and GABAergic afferents to the olfactory bulb in the rat with special emphasis on the projection neurons in the horizontal limb of the diagonal band. J. Comp. Neurul. 243:488-509.

Zilles. K. (1985) The Cortex of the Rat. A Stereotaxic Atlas. Berlin: Springer.

Cortical projection patterns of the medial septum-diagonal band complex

Documents

Transcript of Cortical projection patterns of the medial septum-diagonal band complex