Projections of the Paraventricular and Paratenial Nuclei of the Dorsal Midline Thalamus in the Rat

Home , Epithalamus, Nucleus reuniens, Paratenial nucleus, Suprachiasmatic nucleus, Ventral anterior nucleus

THE JOURNAL OF COMPARATIVE NEUROLOGY 508:212–237 (2008)

ROBERT P. VERTES* AND WALTER B. HOOVER Center for Complex Systems and Brain Sciences, Florida Atlantic University, Boca Raton, Florida 33431

ABSTRACT The paraventricular (PV) and paratenial (PT) nuclei are prominent cell groups of the midline thalamus. To our knowledge, only a single early report has examined PV projections and no previous study has comprehensively analyzed PT projections. By using the anterograde anatomical tracer, Phaseolus vulgaris leucoagglutinin, and the retrograde tracer, FluoroGold, we examined the efferent projections of PV and PT. We showed that the output of PV is virtually directed to a discrete set of limbic forebrain structures, including ‘limbic’ regions of the cortex. These include the infralimbic, prelimbic, dorsal agranular insular, and entorhinal cortices, the ventral subiculum of the hippocampus, dorsal tenia tecta, claustrum, lateral septum, dorsal striatum, nucleus accumbens (core and shell), olfactory tubercle, bed nucleus of stria terminalis (BST), medial, central, cortical, and basal nuclei of amygdala, and the suprachiasmatic, arcuate, and dorsomedial nuclei of the hypothalamus. The posterior PV distributes more heavily than the anterior PV to the dorsal striatum and to the central and basal nuclei of amygdala. PT projections signiﬁcantly overlap with those of PV, with some important differences. PT distributes less heavily than PV to BST and to the amygdala, but much more densely to the medial prefrontal and entorhinal cortices and to the ventral subiculum of hippocampus. As described herein, PV/PT receive a vast array of afferents from the brainstem, hypothalamus, and limbic forebrain, related to arousal and attentive states of the animal, and would appear to channel that information to structures of the limbic forebrain in the selection of appropriate responses to changing environmental conditions. Depending on the speciﬁc complement of emotionally associated information reaching PV/PT at any one time, PV/PT would appear positioned, by actions on the limbic forebrain, to direct behavior toward a particular outcome over a range of outcomes. J. Comp. Neurol. 508: 212–237, 2008. © 2008 Wiley-Liss, Inc.

Indexing terms: medial prefrontal cortex; subiculum of hippocampus; nucleus accumbens; bed nucleus of stria terminalis; central and basal nuclei of amygdala

The paraventricular and paratenial nuclei are promi- global effects on the cortical mantle (Bentivoglio et al., nent cell groups of the midline thalamus (Swanson, 1998; 1991; Groenewegen and Berendse, 1994). The notion, how- Van der Werf et al., 2002). The paraventricular nucleus ever, of the midline thalamus as ‘nonspecific’ has been (PV) lies dorsally on the midline directly below the third revised based on the subsequent anatomical demonstra- ventricle and extends rostrocaudally virtually throughout the thalamus. The paratenial nucleus (PT) borders PV laterally and overlaps with approximately the rostral one- Grant sponsor: National Institute of Mental Health; Grant numbers: third of PV. MH42900, MH63519. Based on the early demonstration that low-frequency *Correspondence to: Dr. Robert P. Vertes, Center for Complex Systems stimulation of the midline and intralaminar nuclei of the and Brain Sciences, Florida Atlantic University, Boca Raton, FL 33431. E-mail: [email protected] thalamus produced slow synchronous activity over wide- Received 29 August 2007; Revised 20 December 2007; Accepted 10 Jan- spread regions of the cortex (recruiting responses) (Demp- uary 2008 sey and Morrison, 1942), the midline thalamus was DOI 10.1002/cne.21679 viewed as ‘nonspecific’ thalamus, exerting nonspecific or Published online in Wiley InterScience (www.interscience.wiley.com).

EFFERENTS OF PV AND PT NUCLEI 213 tion that nuclei of the midline thalamus do not project (mPFC) and ACC were themselves directly connected widely throughout the neocortex, but rather selectively to (mPFC to ACC). With respect to the paraventricular nu- speciﬁc regions of cortex, primarily those of the prefrontal cleus, PV distributes to the prelimbic cortex (of mPFC) cortex (Berendse and Groenewegen, 1991; Van der Werf et and to the medial shell region of ACC, and PL, in turn, al., 2002; Groenewegen and Witter, 2004; Vertes, 2006). In projects to the shell of ACC (Berendse and Groenewegen, addition, recent reports have shown that stimulation of 1990, 1991; Berendse et al., 1992). individual nuclei of the midline thalamus produce selec- Undoubtedly owing to the early emphasis on PV projec- tive effects on their cortical targets—as opposed to wide- tions to the ventral striatum and to mPFC, subsequent spread actions throughout the cortex (Dolleman-Van der reports largely focused on these target sites. Using dual Weel et al., 1997; Bertram and Zhang, 1999; Kung and retrograde labeling techniques, Otake and Nakamura Shyu, 2002; Zhang and Bertram, 2002; Viana Di Prisco (1998) reported that of the nuclei of the midline thalamus, and Vertes, 2006). PV contained the largest percentage of cells with collat- In a series of reports, Groenewegen and colleagues (Be- eral projections to ACC and mPFC. In like manner, Bub- rendse and Groenewegen, 1990, 1991; Berendse et al., ser and Deutch (1998) showed that Ϸ15% of PV cells 1992; Groenewegen et al., 1999) showed that major tar- distribute via collaterals to the medial shell of ACC and PL, gets of midline thalamic nuclei were the prefrontal cortex while Pinto et al. (2003) demonstrated that PV ﬁbers termi- and ventral striatum (nucleus accumbens, ACC), and fur- nate in close apposition to dopaminergic (DA) terminals in ther that recipient zones in the medial prefrontal cortex nucleus accumbens, but not to DA terminals in the mPFC.

Abbreviations

AA Anterior amygdaloid area MFB Medial forebrain bundle ac Anterior commissure MG Medial geniculate nucleus of thalamus AC Anterior cingulate cortex MH Medial habenula ACC,c,s Nucleus accumbens, core and shell divisions MO Medial orbital cortex AGm Medial agranular (frontal) cortex mPFC Medial prefrontal cortex AGl Lateral agranular (frontal) cortex MPO Medial preoptic area AH Anterior nucleus of hypothalamus MPN Medial preoptic nucleus AI,d,p,v Agranular insular cortex, dorsal, posterior, ventral divi- MRF Mesencephalic reticular formation sions MS Medial septum AM Anteromedial nucleus of thalamus mt Mammillothalamic tract AON Anterior olfactory nucleus OC Occipital cortex AV Anteroventral nucleus of thalamus OT Olfactory tubercle BLA Basolateral nucleus of amygdala PAp Posterior parietal cortex BMA Basomedial nucleus of amygdala PC Paracentral nucleus of thalamus BST Bed nucleus of stria terminalis PFC Prefrontal cortex CA1,3 Field CA1 and CA3 of Ammon’s horn PH Posterior nucleus of hypothalamus cc Corpus callosum PHA-L Phaseolus vulgaris-leucoagglutinin CEA,c,l,m Central nucleus of amygdala, capsular, lateral, and me- PIR Piriform cortex dial divisions PL Prelimbic cortex CL Central lateral nucleus of the thalamus PO Posterior nucleus of thalamus CLA Claustrum PRC Perirhinal cortex CM Central medial nucleus of thalamus PT Paratenial nucleus of thalamus COA,a,p Cortical nucleus of amygdala, anterior, posterior divisions PV,a,p Paraventricular nucleus of thalamus, anterior and poste- CP Caudate-putamen rior divisions DBh Nucleus of diagonal band, horizontal limb RE Nucleus reuniens of thalamus DMh Dorsomedial nucleus of hypothalamus RH Rhomboid nucleus of thalamus EC,l Entorhinal cortex, lateral division RN Red nucleus ECT Ectorhinal cortex RSC Retrosplenial cortex EN Endopiriform nucleus RT Reticular nucleus of thalamus fa Forceps of the corpus callosum SC Superior colliculus FG Fluorogold SCN Suprachiasmatic nucleus FI Fimbria of hippocampus SI Substantia innominata FP,l,m Frontal polar cortex, lateral, medial divisions sm Stria medullaris FS Fundus of striatum SM Submedial nucleus of thalamus GI Granular insular cortex SNr Substantia nigra, pars reticulata GP Globus pallidus SPZ Subparaventricular zone of hypothalamus HF Hippocampal formation SSI Primary somatosensory cortex IAM Interanteromedial nucleus of thalamus SSII Secondary somatosensory cortex IL Infralimbic cortex st Stria terminalis IMD Intermediodorsal nucleus of thalamus SUB,v Subiculum, ventral division IP Interpeduncular nucleus SUM Supramammillary nucleus LA Lateral nucleus of amygdala TE Temporal cortex LD Lateral dorsal nucleus of thalamus TT,d,v Tenia tecta, dorsal and ventral divisions LH Lateral habenula V3 Third ventricle LHy Lateral hypothalamus VAL Ventral anterior nucleus of thalamus LO Lateral orbital cortex VB Ventral basal nucleus of thalamus LP Lateral posterior nucleus of thalamus VL Lateral ventricle LPO Lateral preoptic area VLO Ventrolateral orbital cortex LS Lateral septum VM Ventral medial nucleus of thalamus MA Magnocellular preoptic nucleus VO Ventral orbital cortex MD Mediodorsal nucleus of thalamus VTA Ventral tegmental area MEA Medial nucleus of the amygdala ZI Zona incerta The Journal of Comparative Neurology

214 R.P. VERTES AND W.B. HOOVER

To our knowledge, only a single report (Moga et al., ventral striatum, lateral septum, bed nucleus of stria ter- 1995) has examined the general distribution of PV projec- minalis, and to most of the amygdala, with a concentra- tions “with special emphasis on the projections to the tion in the central and basal nuclei of the amygdala. hypothalamus and amygdala.” Focusing on circadian circuitry, Moga et al. (1995) described PV projections to the suprachiasmatic nucleus (SCN) as well as to other sites Materials and Methods involved in circadian rhythms including the dorsomedial Single injections of Phaseolus vulgaris-leucoagglutinin nucleus and subparaventricular zone of the hypothala- (PHA-L) were made into either the PV or PT nuclei of the mus. These results, coupled with the demonstration that midline thalamus in 31 male Sprague–Dawley rats PV receives input from all major components of the circa- (Charles River, Wilmington, MA) weighing 275–400 g. Of dian system including SCN, led Moga et al. (1995) to the 31 injections, 10 were confined to PV, 8 were confined conclude that, “the anterior PV is ideally situated to relay to PT, 4 overlapped PV and PT, 4 overlapped PV and the circadian timing information from the SCN to brain areas mediodorsal nucleus (MD); 3 overlapped PV and the in- involved in visceral and motivation aspects of behavior termediodorsal nucleus (IMD), and 2 were localized to the and to provide feedback regulation of the SCN.” Consis- interanteromedial nucleus (IAM). Another 16 male tent with this, Peng and Bentivoglio (2004) recently dem- Sprague–Dawley rats weighing 350–450 g received single onstrated at the light and electron microscopic (EM) levels injections of the retrograde fluorescent tracer FluoroGold that SCN fibers synaptically contact PV cells projecting to (FG) (Fluorochrome, Denver, CO) into some PV and PT the amygdala and concluded that PV serves an important targets: the central (CEA) and basolateral (BLA) nuclei of role in the “transfer of circadian timing information to the the amygdala and the core and shell of nucleus accum- limbic system.” bens. Of the 16 injections, seven were made in CEA or PV receives a vast array of afferents from monoaminer- BLA of the amygdala and three were control injections in gic and neuropeptide containing systems of the brainstem other nuclei of the amygdala. Of the seven injections in the and hypothalamus—systems known to have activating central and basal nuclei of the amygdala, two were made effects on the forebrain (Chen and Su, 1990; Vertes, 1991; in CEA, two in the basolateral nucleus (BLA), two in BLA Freedman and Cassell, 1994; Bhatnagar et al., 2000; and CEA, and one in BLA and the basomedial nucleus. Of Krout et al., 2002; Otake, 2005). Accordingly, PV (and the six injections in nucleus accumbens, three were local- other midline thalamic nuclei) are thought to serve a ized to the core of ACC and three to the shell of ACC. The direct role in processes of arousal and attention (Krout et experiments were approved by the Florida Atlantic Uni- al., 2002; Van der Werf et al., 2002; Vertes, 2002, 2006). versity Institutional Animal Care and Use Committee and Consistent with a role for PV in attention, Kirouac et al. conform to all federal regulations and National Institutes (2005) recently showed that PV receives pronounced input of Health guidelines for the care and use of laboratory from orexin (hypocretin) containing cells as well as from animals. cocaine- and amphetamine-regulated transcript containing (CART) neurons of the hypothalamus (Kirouac et al., PHA-L procedures 2006), and that both orexin and CART fibers synapse with Powdered lectin from PHA-L was reconstituted to 4–5% PV cells projecting to the shell of ACC (Parsons et al., in 0.05 M sodium phosphate buffer (PB), pH 7.4. The 2006). They proposed that PV links visceral/arousal sys- PHA-L solution was iontophoretically deposited in the tems to limbic forebrain regions involved in behavioral brains of anesthetized rats (sodium pentobarbital, 75 mg/ responses (Parsons et al., 2006). kg, ip) by means of a glass micropipette with an outside tip Taken as a whole, the foregoing suggests that PV may diameter of 40–60 ␮m. Positive direct current (5–10 ␮A) represent an important relay in the transfer of visceral/ was applied through a Grass stimulator (Model 88) cou- arousal, homeostatic, and circadian information to parts pled with a high voltage stimulator (Frederick Haer, Bow- of the limbic system—thereby priming them (state of doinham, ME) at 2 seconds “on” / 2 seconds “off” intervals readiness) for behavioral responding. In this regard, PV for 40–50 minutes. After a survival time of 7–10 days, rats neurons show elevated levels of c-fos expression during were deeply anesthetized with sodium pentobarbital and periods of wakefulness (compared to sleep) (Peng et al., perfused transcardially with a heparinized buffered saline 1995) as well as during stressful conditions (Chastrette et wash (100 mL/animal) followed by a fixative (4% parafor- al., 1991; Bubser and Deutch, 1999; Sica et al., 2000; maldehyde, 0.2–0.5% glutaraldehyde in 0.1 M phosphate Otake et al., 2002)—which could be seen as heightened buffer, pH 7.4) (300–500 mL/animal). The brains were states of arousal. In view, then, of its pivotal role in limbic removed and stored for 2 days at 4°C in 30% sucrose in 0.1 circuitry, we sought to comprehensively examine the ef- M PB. On the following day, 50-␮m frozen sections were ferent projections of PV in the rat. collected in phosphate-buffered saline (PBS, 0.9% sodium Although the paratenial nucleus of the thalamus also chloride in 0.01 M sodium phosphate buffer, pH 7.4) using appears to receive a vast array of afferent information a sliding microtome. Six series of sections were taken. A (Chen and Su, 1990; Krout et al., 2002) and may selec- complete series of sections was treated with 1% sodium tively target structures of the limbic forebrain (Kelley and borohydride in 0.1 M PB for 30 minutes to remove excess Stinus, 1984; Carlsen and Heimer, 1986), little is known reactive aldehydes. Sections were then rinsed in 0.1 M PB, regarding the connections of PT. The purpose, then, of the followed by 0.1 M Tris-buffered saline (TBS), pH 7.6. Fol- present study was to analyze, compare, and contrast ef- lowing this, sections were incubated for 60 minutes at ferent projections of PV and PT nuclei of the midline room temperature (RT) in 0.5% bovine serum albumin thalamus. We show that, with some important differences, (BSA) in TBS to minimize nonspecific labeling. The sec- the output of both PV and PT is mainly directed to the tions were then incubated overnight at RT in diluent medial prefrontal and entorhinal cortices, the ventral sub- (0.1% BSA in TBS containing 0.25% Triton X-100) and iculum of the hippocampus, claustrum, the dorsal and biotinylated goat anti PHA-L (Vector Labs, Burlingame, The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 215

CA) at a concentration 1:500. Sections were then washed TABLE 1. Relative Density of Anterogradely Labeled Fibers Produced by in0.1MPB(4ϫ 8 minutes) and placed in a 1:500 con- PHA-L Injections into PVa, PVp, and PT of the Midline Thalamus centration of biotinylated rabbit antigoat immunoglobulin Labeling (IgG) and diluent for 2 hours. Sections were washed and Structure PVa PVp PT then incubated in a 1:100 concentration of peroxidase- avidin complex from the Elite kit (Vector) and diluent for Telencephalon Cortex 1 hour. Following another 0.1 M PB wash the peroxidase Agranular insular, dorsal ϩϩ ϩϩ ϩϩϩ reaction product was visualized by incubation in a solu- Agranular insular, posterior ϩϩϩϩ Ј Agranular insular, ventral ϪϪϪ tion containing 0.022% 3,3 diaminobenzidine (DAB, Al- Agranular, lateral (primary motor) ϪϪϪ ϪϪϪ drich, Milwaukee, WI) and 0.003% H2O2 in TBS for 6 Agranular, medial (secondary motor) ϫ Anterior cingulate, dorsal Ϫϩϩϩϩ minutes. Sections were then rinsed again in PBS (3 1 Anterior cingulate, ventral Ϫϩϩϩ minutes) and mounted onto chrome-alum gelatin-coated Ectorhinal ϩϩϩϩ slides. An adjacent series of sections from each rat was Entorhinal ϩ ϩϩ ϩϩϩ Frontal polar, lateral ϩϩϪ stained with cresyl violet for anatomical reference. Sec- Frontal polar, medial ϩϩϩϩϩ tions were examined using light and darkfield optics. In- Granular insular ϪϩϩϪ Infralimbic ϩ ϩϩϩ ϩϩϩ jection sites, cells, and labeled fibers were plotted on rep- Lateral orbital ϪϪϪ resentative schematic coronal sections through the brain Medial orbital ϩϩϩϩϩ Parietal ϪϪϪ using sections adapted from the rat atlas of Swanson Perirhinal ϩϩϩϩϩ (1998). Brightfield and darkfield photomicrographs of in- Piriform ϩϩϩ jection sites and labeled fibers were taken with a Nikon Prelimbic ϩϩ ϩϩϩ ϩϩϩ Retrosplenial ϪϪϩ DXM1200 camera mounted on a Nikon Eclipse E600 mi- Subiculum, ventral ϩϩ ϩϩ ϩϩ croscope. Digital images were captured and reconstructed Somatosensory, secondary ϪϪϪ Ventral Orbital ϩϩ Ϫ ϩϩ using Image-Pro Plus 4.5 (Media Cybernetics, Silver Ventrolateral Orbital ϪϪϪ Springs, MD), and adjusted for brightness and contrast Accumbens n. Core ϩϩϩ ϩϩ ϩϩϩ using Adobe PhotoShop 7.0 (Mountain View, CA). Pat- Shell ϩϩϩ ϩϩϩ ϩϩϩ terns of labeling were classified as light, moderate, and Amygdala dense (Table 1), with ‘light’ referring to a few labeled Anterior amygdaloid area ϩϩϩϩ Basolateral n. ϩ ϩϩ ϩϩϩ fibers widely dispersed throughout a structure, ‘dense’ as Basomedial n. ϩϩ ϩϩϩ ϩϩϩ a heavy concentration of labeled fibers generally occupy- Central n. ϩϩϩ ϩϩϩ ϩϩ Cortical n. Ϫϩϩϩϩ ing a significant portion (or most) of a structure, and Lateral n. Ϫ ϩϩ ϩϩϩ ‘moderate’ between these two patterns. Medial n. Ϫ ϩϩϩ ϩϩ Posterior n. ϪϪϩϩ FluoroGold procedures Amygdaloid-piriform area ϪϪϩϩ Bed n. of stria terminalis ϩϩϩ ϩϩϩ ϩϩϩ FluoroGold (FG) (Fluorochrome) was dissolved in a 0.1 Claustrum ϩϩ ϩϩ ϩϩ Diagonal band of Broca, horizontal ϪϪϪ M sodium acetate buffer (pH 4.0 to 5) to yield a 4–5% Endopiriform n. ϪϪϪ concentration. The FG solution was iontophoretically de- Globus pallidus ϩϩϪ Medial preoptic area Ϫϩϩϩ posited in the brains of anesthetized rats by means of a Olfactory tubrical ϩϩ ϩϩ ϩϩϩ glass micropipette with an outside tip diameter of 25–50 Septum ␮ Medial ϪϪϪ m. Single FG injections were made into one of four struc- Lateral ϩ ϩϩ ϩϩϩ tures in separate rats: the central and basolateral nuclei Striatum ϩϩ ϩϩϩ ϩϩϩ of the amygdala and the core and shell of ACC. The pro- Substantia innominata ϪϩϪ Tenia tecta ϩϩϩϩϩ cedures for FG injections were basically the same as de- Diencephalon scribed for PHA-L injections, with the following excep- Hypothalamus Arcuate n. ϩϩ Ϫ Ϫ tions: 1) the outside tip diameter of the glass Dorsomedial n. ϩϩϩϪ micropipettes was 25–50 ␮m, and 2) the length of injec- Lateral hypothalamic area ϪϪϩϩ Paraventricular n. ϪϪϪ tions was 2–10 minutes. Following a survival time of 7 Posterior hypothalamus ϪϪϪ days, rats were deeply anesthetized with sodium pento- Suprachiasmatic n. ϩϩ ϩϩ Ϫ barbital and perfused transcardially with 100 mL of a Supramammillary n. ϪϪϪ Thalamus heparinized saline wash followed by 450 mL of fixative Central medial n. ϪϪϪ (4% paraformaldehyde in 0.01 M sodium PB, pH 7.4). The Interanteromedial n. ϪϪϪ Mediodorsal n. ϪϪϪ brains were then removed and stored for 48 hours in a Parafascicular n. ϪϪϪ sucrose solution (30% sucrose in 0.1 M PB) at 4°C. Follow- Parataenial n. ϪϪϪ ␮ Paraventricular n. ϩϪϪ ing this, 50- m coronal sections were taken on a freezing Reticular n. ϪϪϪ microtome and collected in 0.1 M PB and stored at 4°C. Six Reuniens n. Ϫϩϩ series of sections were taken. A complete series of sections Rhomboid n. ϪϩϪ was incubated in a sodium borohydride solution (1% so- ϩ, light labeling; ϩϩ, moderate labeling; ϩϩϩ, dense labeling; Ϫ, absence of labeling; dium borohydride in 0.1 M PB) for 30 minutes, and n, nucleus washed with 0.1M PB four times at 6 minutes each (4 ϫ 6 min). The sections were then blocked in a Tris-saline solution (0.5% BSA, Sigma Chemicals, St. Louis, MO; 0.25% Triton X-100 in 0.1 M Tris-saline, pH 7.6) for 1 Following incubation in the primary antiserum, sections hour. Following the blocking procedure the sections were were washed (4 ϫ 6 min) in 0.1 M PB, then incubated in a incubated for 48 hours at RT in a primary antiserum secondary antiserum (biotinylated goat antirabbit IgG) directed against FG (rabbit anti-FluoroGold) (Chemicon, (Vector) at a concentration of 1:500 in diluent for 2 hours. Temecula, CA) at a concentration of 1:1,000 in diluent. Sections were then washed again (4 ϫ 6 minutes) and The Journal of Comparative Neurology

216 R.P. VERTES AND W.B. HOOVER

Fig. 1. High-power brightfield photomicrographs at two levels of clearly visible PHA-L filled cells in PVa (B), PVp (D), and PT (F). Scale magnification of PHA-L injections sites in the anterior paraventricu- bar ϭ 375 ␮m for A,E; 200 ␮m for B; 400 ␮m for C; 300 ␮m for D; 225 lar nucleus (A,B), the posterior paraventricular nucleus (C,D), and ␮m for F. the paratenial nucleus (E,F) of the dorsal midline thalamus. Note The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 217

Fig. 2. Schematic representation of labeling present in select sections through the forebrain and midbrain (A–N) produced by a PHA-L injection (dots in I,J) in the anterior part of the paraventricular nucleus of the thalamus (case 6). Sections modiﬁed from the rat atlas of Swanson (1998). See list for abbreviations.

incubated in avidin-biotin complex (Vector) at a 1:100 H2O2 in TBS for 6 minutes. Sections were rinsed again in concentration in diluent for 1 hour. After a ﬁnal set of 4 ϫ PBS (3 ϫ 1 minutes) and mounted onto chrome-alum 6 minute rinses the peroxidase reaction product was visu- gelatin-coated slides. An adjacent series of representative alized by incubation in a solution containing 0.022% of sections from each rat was stained with cresyl violet for DAB (Aldrich), 0.015% nickel chloride (NiCl), and 0.003% anatomical reference. The resulting material was pro- The Journal of Comparative Neurology

218 R.P. VERTES AND W.B. HOOVER

Figure 2 (Continued)

cessed for presentation as described for the PHA-L sec- Anterior paraventricular nucleus (PVa) tions. (case 6) Figure 2 schematically depicts the pattern of distribu- Results tion of labeled fibers following a PHA-L injection in the The pattern of distribution of projections from the PV anterior part of PV (case 6). As shown, labeled fibers and PT nuclei of the thalamus are described. Figure 1 coursed ventrolaterally from the site of injection (Fig. 2F) shows sites of injection in the anterior PV (PVa) (Fig. within the thalamic peduncle to regions of the lateral 1A,B), the posterior PV (PVp) (Fig. 1C,D), and PT (Fig. hypothalamus and from there took three principal routes. 1E,F) at two levels of magnification. As depicted, PHA-L- A major contingent continued ventrolaterally in amyg- filled cells are restricted to respective nuclei. The patterns dalopetal pathways to reach the amygdala and surround- of labeling obtained with the schematically depicted cases ing regions of cortex, others coursed rostrally to the ante- are representative of patterns seen with nonillustrated rior forebrain primarily bound for the ventral striatum cases. and the prefrontal cortex or caudally en route to regions of The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 219 the hypothalamus. Some labeled fibers of the ascending Posterior paraventricular nucleus (PVp) bundle joined the stria terminalis and traveled with it to (case 32) reach the amygdala and adjacent regions of cortex. Anterior and posterior PV injections largely gave rise to Overall, labeling was light at the anterior pole of the similar patterns of labeling but, as described below, over- forebrain (Fig. 2A,B). A small collection of labeled fibers all density of labeling was stronger with PVp than with was seen along the medial wall of the prefrontal cortex PVa injections. (PFC), mainly within the anterior prelimbic (PL) cortex. Labeled fibers from PVp mainly coursed rostrally Some diffuse labeling was also observed in the dorsal through the dorsal thalamus (Fig. 4K–N) and approxi- tenia tecta (TTd) (Fig. 2A,B). mately at the level of the rostral pole of the hippocampus Further caudally at the rostral forebrain (Fig. 2C,D), (Fig. 4I,J) turned ventrolaterally to exit the thalamus. labeling was primarily confined to PL of the mPFC and to From there, they either continued on the same trajectory rostral aspects of nucleus accumbens (ACC). As depicted to reach the amygdala and surrounding regions of the in Figure 3A, labeled fibers mainly encircled the outer dorsal striatum and cortex or ascended or descended boundaries of ACC, and were much less concentrated in through the medial forebrain bundle (MFB) en route to the core of the rostral ACC. Additional labeled sites were the basal forebrain and prefrontal cortices, rostrally, or to the anterior claustrum (CLA), ventromedial regions of the parts of the hypothalamus, caudally. dorsal striatum bordering ACC (Fig. 2D), and to a lesser Similar to PVa, labeling at the anterior pole of the degree the dorsal agranular insular cortex (AId) (Fig. 2C). forebrain (Fig. 4A,B) was generally moderate and mainly There was a noticeable lack of labeling at this level (Fig. present in inner layers of the anterior PL and anterior 2C), as well as caudally, over most of the cortical mantle. cingulate (AC) cortices and to considerably lesser degrees Labeling was stronger on the left than on the right side of in medial frontal polar (FPm), medial orbital (MO), and the brain (Fig. 2A–M), reflecting the fact that the PV dorsal agranular insular (AId) cortices. Moderate num- injection was slightly lateralized to the left side (Fig. bers of labeled axons were also visible in TTd. 1A,B). Further caudally at the rostral forebrain (Fig. 4C), la- At septal levels (Fig. 2D–H), labeling was pronounced beled fibers spread widely over ventral aspects of the brain mainly localized to the ventral mPFC, claustrum, dorsal and relatively restricted to the core and shell regions of agranular insular cortex (AId), rostral ACC, and the ol- ACC, to ventral aspects of the lateral septum (LS) (Figs. factory tubercle (OT). As depicted (Fig. 4C), labeling was 2F, 3B), and to the bed nucleus of the stria terminalis quite dense in the inner layers of the infralimbic (IL) and (BST). The dense labeling within the shell of ACC and to prelimbic cortices and somewhat less pronounced in AId, a slightly lesser degree in the core of ACC is depicted in CLA, rostral ACC, and OT. A few labeled fibers were also the photomicrographs of Figure 3B, while equally pro- seen in AC. nounced labeling of BST, above and below the anterior The main targets of labeled fibers further caudally in commissure, is shown in Figure 3C. Outside of these sites, the forebrain were the dorsal and ventral striatum (Fig. CLA, the olfactory tubercle (OT), and ventral regions of 4D–F). As depicted schematically (Fig. 4D,E) and in the the dorsal striatum were moderately labeled, while the micrograph of Figure 5A, the shell of ACC (ACCs) was ventral globus pallidus (Fig. 2H) was lightly labeled. intensely labeled. With the exception of the region sur- At mid levels of the forebrain (Fig. 2I–K), labeling was rounding the anterior commissure, which was heavily la- mainly confined to the amygdala and parts of the hypo- beled, the core of ACC (ACCc) was moderately labeled. thalamus. The arcuate and suprachiasmatic (SCN) (Fig. Unlike PVa, significant numbers of labeled axons were 2I) nuclei of the hypothalamus were moderately labeled. also present in ventral aspects of the dorsal striatum (CP), Within the (rostral) amygdala, labeling was very heavy progressively thinning from the ventrolateral to dorsome- within the central nucleus (CEA) (Fig. 2J,K), particularly dial CP. Other moderately to heavily labeled sites at these within the lateral CEA (Fig. 3D), moderately dense in the levels were ventral LS, OT, CLA, and AId (Fig. 4D–F). basomedial (BMA) and basolateral (BLA) nuclei, and rel- Immediately caudal to ACC (Fig. 4F–H), labeled axons atively light in the medial and cortical nuclei of amygdala. spread densely throughout the extent of the bed nucleus of Some labeled fibers were also present on the lateral con- the stria terminalis (BST) and were also present in size- vexity of cortex within the posterior agranular insular able numbers in medial aspects of CP, CLA, OT, AId, and cortex, rostrally (Fig. 2I,J) and in the perirhinal (PRC), the suprachiasmatic nucleus (SCN). Figure 5B shows a rostral entorhinal (EC) and piriform cortices, caudally discrete group of labeled axons bilaterally within SCN. Additional light to moderately labeled sites included the (Fig. 2K). posterior agranular insular cortex (AIp) (with some exten- At further caudal levels of the forebrain (Fig. 2L,M) and sion dorsally into the granular insular cortex), substantia the rostral midbrain (Fig. 2N), the bulk of labeled fibers innominata (SI), medial preoptic area (MPO) and the glo- was localized to the amygdala and surrounding regions bus pallidus (GP) (Fig. 4F–H). including caudal parts of the dorsal striatum (Fig. 2L,M), At mid to caudal levels of the forebrain (Fig. 4I–N), deep layers of the perirhinal, entorhinal and piriform cor- labeled fibers were mainly confined to the dorsal stria- tices (Fig. 2L–N), and the ventral subiculum of the hip- tum and to the amygdala, spreading massively through- pocampus (Fig. 2N). Within the amygdala, labeling was out the amygdala. As shown (Figs. 4I–N, 6A–D), labeled predominantly restricted to the basal nuclei—dense fibers virtually blanketed the amygdala, with the dens- within BMA and moderate within the medial part of BLA. est concentration in the central (CEA) (Fig. 6A,B) and Some labeled fibers were also present in the posterior PV basomedial (BMA) (Fig. 6A–D) nuclei of amygdala. The (PVp) and in the dorsomedial nucleus of the hypothala- medial (MEA) and basolateral nuclei of amygdala were mus (Fig. 2L,M). also fairly heavily labeled, whereas the lateral and The Journal of Comparative Neurology

220 R.P. VERTES AND W.B. HOOVER

Fig. 3. A–D: Low-magnification darkfield photomicrographs of labeled fibers distribute massively to caudal part of ACC. Note pro- transverse sections through anterior (A–C) and posterior (D) regions nounced labeling in both the shell and core of ACC, with additional of the forebrain depicting patterns of labeling in rostral and caudal labeling in the adjacent lateral septum (LS) and ventromedial parts of nucleus accumbens (ACC) (A,B), the bed nucleus of the stria termi- the dorsal striatum (caudate-putamen) (CP). C: Note strong terminal nalis (C), and the amygdala (D) produced by a PHA-L injection into labeling in BST above and below the anterior commissure. D: Note a anterior paraventricular nucleus of thalamus. A: Note that labeled dense aggregate of labeled fibers in the central medial and medial fibers mainly encircle but largely avoid the central core of the rostral part of the medial part of the basomedial nucleus and prominent but ACC, and also note significant labeling in the prelimbic cortex (PL) of less dense labeling in the basolateral nucleus of the amygdala. Scale the medial prefrontal cortex. B: By contrast with the rostral ACC (A), bar ϭ 550 ␮m for A; 500 ␮m for B–D. See list for abbreviations.

parts of the anterior and posterior cortical nuclei of dial CP, abutting the globus pallidus, was observed. amygdala were moderately labeled (Fig. 4I–N). At pre Labeling was densest ventrally in medial CP (Fig. and beginning levels of the hippocampus (Fig. 4H–K), a 4I–M) at its point of merger with medial aspects of the relatively narrow band of labeled ﬁbers within the me- amygdala. The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 221

Fig. 4. Schematic representation of labeling present in select sections through the forebrain and midbrain (A–O) produced by a PHA-L injection (dots in N) in the posterior part of the paraventricular nucleus of the thalamus (case 32). Sections modiﬁed from the rat atlas of Swanson (1998). See list for abbreviations.

Further caudally (Fig. 4L–N), labeled ﬁbers were found (BMA and BLA), and cortical nuclei of amygdala were to extend laterally from the medial CP to occupy most of moderately to densely labeled (Fig. 4L–N). Signiﬁcant the mediolateral expanse of the caudal CP. In addition, as numbers of labeled axons were also visible within inner rostrally, posterior parts of the lateral, central, basal layers of the parahippocampal and piriform cortices; that The Journal of Comparative Neurology

222 R.P. VERTES AND W.B. HOOVER

Figure 4 (Continued) The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 223 is, a continuous column in layers 5 and 6 stretching from moderate numbers were present in the dorsomedial nu- the ectorhinal, perirhinal, and entorhinal cortices to the cleus of the hypothalamus (DMH) (Fig. 5C) and a few piriform cortex (Fig. 4L–N). Although most regions of the within the midline thalamus–reuniens (RE) and rhomboid diencephalon were devoid of labeled fibers (Fig. 4L–N), (RH) nuclei. At the level of the midbrain (Fig. 4O) moderate numbers of labeled fibers were present in the caudal perirhinal cortex, lateral EC, and the ventral subiculum of the ventral hippocampus. Although labeling progressively thinned caudally, labeled axons continued to be present in lateral EC and the ventral subiculum throughout caudal reaches of the brain (not shown). With a few exceptions, there was a general absence of labeling at the rostral midbrain (Fig. 4O) and throughout the brainstem. Paratenial nucleus (PT) (case 27) As depicted (Fig 1E,F), the injection in the paratenial nucleus (PT) was confined to left PT, and accordingly labeling was virtually restricted to the left side of the brain. Labeling was minimal contralateral to the injection. Similar to PV, labeled fibers from PT exited ventrolaterally through the thalamus and then either continued on the same course to the amygdala or ascended through the MFB to the anterior forebrain or descended in the MFB to parts of the caudal diencephalon. At anterior pole of the forebrain (Fig. 7A,B), virtually the entire medial wall of the mPFC was densely labeled. This includes the medial frontal polar, prelimbic (PL), and medial orbital (MO) cortices, rostrally (Fig. 7A) and the anterior cingulate, PL, and MO cortices, further caudally (Fig. 7B). This is shown in the photomicrographs of Figure 8A,B. As depicted, labeled fibers spread to all layers of respective cortices, but were most heavily concentrated in layers 2/3 of PL and the ventrally adjacent MO. In addition, moderate numbers of labeled fibers were present in the ventral orbital cortex (VO) and TTd, but considerably fewer in AId. Further caudally in the anterior forebrain (Fig. 7C), labeling remained pronounced along the medial wall of the ventral mPFC, particularly pronounced in layers 1/3 of the infralimbic (IL) and prelimbic cortices (Figs. 7C, 8C, 9). Equally dense labeling was observed within the rostral ACC and parts of OT. Additional lightly to moderately labeled sites included AId, AC, and the CLA. This pattern of labeling is depicted in the brightfield photomicrograph of Figure 9. At early septal levels (Fig. 7D–F), labeled axons were primarily localized to the ventral mPFC, CP, ventral striatum, CLA, and AId. Labeling was pronounced (or massive) in IL and PL of the mPFC, ventromedial sectors of CP, the core and shell of ACC, and parts of OT (Fig. 7D–F). Some regions of AC, AId, and LS were also heavily labeled (Fig. 7E,F). As depicted schematically (Fig. 7D–F) and in

Fig. 5. A–C: Low-magniﬁcation darkﬁeld photomicrographs of transverse sections through the forebrain depicting patterns of labeling produced by a PHA-L injection into posterior paraventricular nucleus. A: Note the massive labeling throughout the shell division of the nucleus accumbens (ACCs) and intense but lesser labeling in the core of ACC (ACCc) surrounding the anterior commissure. B: Note the strong labeling bilaterally in the suprachiasmatic nucleus of the hypothalamus (SCN) above the optic chiasm. C: Note the pronounced labeling bilaterally in the dorsomedial nucleus (DMh) of the hypothalamus lateral to the third ventricle. Scale bar ϭ 500 ␮m for A; 250 ␮m for B; 300 ␮m for C. See list for abbreviations. The Journal of Comparative Neurology

224 R.P. VERTES AND W.B. HOOVER

Fig. 6. A–D: Series of low-magnification darkfield photomicro- and basolateral nuclei of amygdala, and prominent but less dense graphs of transverse sections rostrocaudally through the forebrain labeling in the parts of the medial (MEA), lateral (LA), and anterior (A–D) depicting patterns of labeling within the amygdala produced by cortical nuclei of amygdala. C,D: Note labeling at caudal levels of the a PHA-L injection into posterior paraventricular nucleus. A,B: Note amygdala mainly confined to BMA and BLA. Scale bar ϭ 750 ␮m. See very intense dense labeling in the central (CEA), basomedial (BMA) list for abbreviations.

the photomicrograph of Figure 10A, labeling was uneven observed in the rostral amygdala and adjoining regions of mediolaterally across ACC and within bordering regions the piriform cortex. As shown (Fig. 7I–K), labeled fibers of CP; that is, dense in the internal (medial) shell of ACC, spread widely (and moderately) throughout the amygdala considerably lighter in the lateral shell (with some rela- to the anterior amygdaloid area (AA), CEA, MEA, BLA, tively clear pockets), very strong in the (dorsal) core of BMA, and the anterior cortical nucleus (COAa). There was ACC, extending into the ventromedial CP, and moderate a noticeable absence of labeling in the core of CEA (Fig in the lateral CP. 7J,K). Other lightly to moderately labeled sites were AC, At caudal levels of the septum (Fig. 7G,H) significant dorsomedial CP, CLA, and anterior and lateral nuclei of numbers of labeled axons were present in rostral BST, on the hypothalamus. the lateral and ventral border of the anterior commissure, At mid-levels of the forebrain (Fig. 7L–N), labeled and medially within CP abutting lateral wall of the lateral fibers were essentially confined to the ventrolateral sec- ventricle, dorsal to BST. Light to moderate labeling was tor of the brain; that is, to CP, to the amygdala, and to also observed in AC, LS, ventromedial CP, olfactory tu- the perirhinal, entorhinal, and piriform cortices. An bercle, and CLA. intensely labeled band of tissue stretching diagonally There was marked decline in labeling further caudally through the amygdala was observed (Fig. 7L–N, 10B) in the forebrain (Fig. 7I–K) with labeled fibers mainly that included medial aspects of the lateral and basolat- The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 225

Fig. 7. Schematic representation of labeling present in select sections through the forebrain and midbrain (A–O) produced by a PHA-L injection (dots in I,J) in the paratenial nucleus of the thalamus (case 27). Sections modiﬁed from the rat atlas of Swanson (1998). See list for abbreviations.

eral amygdala and virtually the extent of posterior amygdaloid-piriform area, and posterior cortical nuclei BMA. As rostrally, the core of CEA was largely devoid of of the amygdala (Fig. 7L–N). A few labeled ﬁbers were labeled ﬁbers (Fig. 7L), but moderate numbers were also present in RE, the zona incerta (ZI), and through- seen in the capsular CEA as well as in the posterior, out the lateral hypothalamus. The Journal of Comparative Neurology

226 R.P. VERTES AND W.B. HOOVER

Figure 7 (Continued) The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 227

Fig. 8. A–C: Series of rostrocaudally (A–C) aligned low- intense labeling along the ventral medial wall of the mPFC mainly magnification darkfield photomicrographs of transverse sections confined to the prelimbic (PL) (A–C), medial orbital (MO) (A,B), and through the anterior forebrain depicting patterns of labeling within infralimbic (C) (IL) cortices. As depicted, labeling was particularly the medial prefrontal cortex (mPFC) produced by a PHA-L injection dense in layers 1 and 3 of these prefrontal fields. Scale bar ϭ 750 ␮m. into the paratenial nucleus of the thalamus. Note the presence of

At the rostral midbrain (Fig. 7O), labeled fibers con- ACCs injections; and 3) labeling was slightly stronger in tinued to mainly occupy ventrolateral regions of the PVp with ACCc than ACCs injections. This supports an- brain localized to perirhinal cortex, lateral EC, and the terograde results showing pronounced terminal labeling ventral subiculum of the hippocampus. This pattern of in the shell and core of ACC with PVa, PVp, and PT labeling is depicted in the micrographs of Figure 11A–C injections. showing prominent labeling in these areas— Figure 13 shows FG injections in the rostral BLA (Fig. particularly dense rostrocaudally throughout the lat- 13A) and rostral CEA (Fig. 13D), together with patterns eral EC. The few labeled fibers present dorsally in the of cell labeling in PVa and PT (Fig. 13B,E) and PVp (Fig. retrosplenial cortex (Fig. 7O) mainly appeared bound 13C,F) produced by these injections. As depicted, there for the dorsal subiculum which was lightly labeled is a small number of labeled neurons in PT (Fig. 13B) (Fig. 7O). with the BLA injection and fewer still in PT (Fig. 13E) Retrograde tracing experiments with the CEA injection. This is consistent with anterograde findings showing light terminal labeling in rostral Two major destinations of labeled fibers of PVa, PVp, BLA (Fig. 7I,J) and general lack of labeling in rostral and PT were the nucleus accumbens (shell and core) and CEA (Fig. 7I,J) with PT injections. FG injections in BLA the amygdala—mainly CEA and basal nuclei (see above). To confirm anterograde findings and provide further in- gave rise to pronounced cell labeling in PVp (Fig. 13C), but formation on the distribution of PV and PT fibers to these light labeling in PVa (Fig. 13B), while those in CEA pro- sites, retrograde injections (FG) were made in ACC and duced significant labeling in both PVa (Fig. 13E) and PVp the amygdala and patterns of labeling in PV and PT (Fig. 13F). This is consistent with anterograde results dem- determined. onstrating strong terminal labeling in rostral CEA with PVa Figure 12 shows FG injections in the shell (Fig. 12A) (Fig. 2I,J) and with PVp injections (Figs. 4I,J, 6A,B), as well and the core (Fig. 12D) of ACC together with patterns of as weak labeling in BLA with PVa injections (Fig. 2I,J) and cell labeling in PVa and PT (Fig. 12B,E) and PVp (Fig. dense labeling in BLA (Figs. 4I,J, 6A) with PVp injections. 12C,F) obtained with these injections. As depicted: 1) la- As depicted, labeled cells are also present in RE (Fig. 13B,E), beled cells were present in PVa, PVp, and densely in PT the intermediodorsal (IMD), and the central medial (CM) with injections in the shell (ACCs) and core (ACCc) of nuclei of the thalamus (Fig. 13C,F) with BLA and CEA ACC; 2) labeling was heavier in PVa with ACCc than injections. The Journal of Comparative Neurology

228 R.P. VERTES AND W.B. HOOVER

(LS), the core and shell of ACC, OT, BST, several nuclei of the amygdala, including the lateral, medial, central, basal (BLA and BMA), and the anterior and posterior cortical nuclei, and the suprachiasmatic (SCN), arcuate, and dorsomedial nuclei of the hypothalamus. Secondary targets were the anterior cingulate and ectorhinal cortices, dorsal tenia tecta (TTd), the medial preoptic area, reuniens (RE), and rhomboid nuclei of the thalamus and the lateral hypothalamus. There is a significant overlap in projections from the anterior (PVa) and posterior (PVp) PV (Figs. 2, 4; Table 1). With some exceptions, PVp is the source of stronger projections to most commonly innervated sites. Perhaps the most significant difference between PVa and PVp projections is that PVa distributes minimally to the dorsal striatum (caudate-putamen), whereas PVp projects quite massively to CP, mainly to medial/ventromedial regions of CP. In addition, while PVa and PVp project commonly to the amygdala, PVp distributes more widely and heavily throughout the amygdala than PVa, particularly to the basal nuclei of amygdala. On the other hand, PVa is the source stronger projections to the ventral subiculum of the hippocampus. Overview of PT projections and comparisons with PV projections The main targets of PT were the medial frontal polar (FPm), anterior cingulate, prelimbic, infralimbic, medial orbital, dorsal agranular insular, piriform and entorhinal cortices, the ventral subiculum of hippocampus, the claustrum, the core and shell of nucleus accumbens, the medial striatum (CP), BST, and caudal parts of the central and Fig. 9. Low-magnification brightfield photomicrograph of a transverse section through the anterior forebrain depicting patterns of basal nuclei of amygdala. PT also distributes to the ven- labeling within the anterior cingulate (AC), prelimbic (PL), infralim- tral orbital and perirhinal cortices, the dorsal subiculum bic (IL), and dorsal agranular insular cortices, the olfactory tubercle of hippocampus, lateral septum, olfactory tubercle, medial (OT), and the rostral pole of the nucleus accumbens (ACC) produced and cortical nuclei of amygdala, RE of thalamus, and the by a PHA-L injection in the paratenial nucleus of thalamus. Scale lateral hypothalamus. bar ϭ 500 ␮m. See list for abbreviations. Although there is considerable overlap in PT and PV projections, there are several important differences between the two sets of projections. PT sends considerably Discussion stronger projections than PV to the mPFC, to the lateral We examined, compared, and contrasted the efferent entorhinal cortex, to the ventral subiculum, and to ante- projections of the PV and PT nuclei of the dorsal midline rior regions of the dorsal and ventral striatum. Differences thalamus in the rat. The main (or virtually sole) targets of are particularly notable with respect to the mPFC. PT PV and PT were ‘limbic/limbic related’ structures of the strongly targets the mPFC, distributing throughout the forebrain. With the possible exception of the piriform cor- ventral mPFC to the medial frontal polar, medial orbital tex, there was essentially lack of PV/PT projections to (MO), anterior cingulate, prelimbic and infralimbic corti- ‘nonlimbic’ regions of the cortex including sensorimotor, ces, and particularly heavily in outer layers (1 and 3) of special sensory, or associational cortices as well as few MO, PL, and IL (Fig. 8A–C). By contrast, the projections of projections to most of the thalamus and hypothalamus. As PV (PVa and PVp) to the mPFC are modest and mainly developed below, based on widespread afferents from the confined to IL and PL. With respect to nucleus accumbens brainstem and hypothalamus coupled with output to se- (ACC), PT distributes more heavily to the rostral pole (Fig. lect structures of the limbic forebrain, PV/PT appear crit- 9) and core of ACC (Figs. 10A, 12E), but less densely to the ical for routing visceral/emotional information to struc- shell of ACC (Figs. 3B, 5A) than does PV. Unlike PVa, but tures of the limbic forebrain, including the limbic cortex, similar to PVp, PT strongly targets the dorsal striatum. in the control of goal-directed behaviors. PT fibers terminate densely (and selectively) in the ros- tromedial CP, dorsal to the core of ACC (Fig. 10A), while Overview of PV projections and comparisons PVp distributes rostrocaudally throughout CP and heavily between PVa and PVp projections to the caudal CP—a region basically devoid of fibers from The main targets of PV (anterior and posterior parts) PT. Finally, in contrast to robust PV projections to virtu- were the prelimbic (PL), dorsal agranular insular, perirhi- ally the entire amygdala, PT distributes significantly to nal (PRC) and entorhinal cortices, the ventral subiculum caudal, but at best modestly, to rostral parts of the amyg- of the hippocampus, the claustrum, the lateral septum dala (see Fig. 13B,E). The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 229

Fig. 10. A,B: Low-magniﬁcation darkﬁeld photomicrographs of missure) with a continuation of equally dense labeling into ventrome- transverse sections through the forebrain depicting patterns of label- dial parts of the dorsal striatum (caudate putamen, CP). B: Note ing the dorsal and ventral striatum (A) and the amygdala (B) pro- heavy labeling in parts of the lateral and basolateral nuclei, some- duced by a PHA-L injection into the paratenial nucleus of the thala- what less pronounced labeling in the basomedial nucleus and an mus. A: Note the dense labeling, but uneven labeling (sparsely labeled absence of labeling in the lateral part of the central nucleus of amyg- pockets) in the shell of nucleus accumbens (ACC) and the massive dala. Scale bar ϭ 550 ␮m for A; 500 ␮m for B. See list for abbrevia- labeling in the core of ACC (dorsal/dorsomedial to the anterior com- tions. The Journal of Comparative Neurology

230 R.P. VERTES AND W.B. HOOVER

considerably for the posterior PV. In accord with Moga et al. (1995), we demonstrated pronounced PVa projections to the shell of ACC and to the central and basomedial nuclei of amygdala, but unlike them, also described sub- stantial PVa projections to the core of ACC as well as to the medial, basolateral, and cortical nuclei of amygdala. On the other hand, they demonstrated denser projections to nuclei of the hypothalamus including the retrochias- matic nucleus, subparaventricular zone, and the ventromedial nucleus of the hypothalamus. With respect to PVp, however, Moga et al. (1995) reported that PVp projections were much lighter overall than PVa projections, while we generally found the opposite: stronger PVp than PVa projections to most sites. Further, Moga et al. (1995) described an essential lack of PVp projections to several sites in which we observed them including the infralimbic, piriform, perirhinal and agranular insular cortices, the ventral subiculum, medial regions of the striatum, posterior BLA and BMA, and most of the hypothalamus. The reasons for these differences are unclear but could involve differences in size and locations of the PVp injections. PV projections to emotional/visceral associated forebrain areas. PV distributes to several forebrain sites associated with emotional behavior including the infralimbic cortex, the lateral septum, bed nucleus of stria terminalis, and almost the entire amygdala—with massive projections to CEA. In accord with present findings, previous reports, using various tracers, have described significant PV projections to the infralimbic cortex in rats (Berendse and Groenewe- gen, 1991; Conde et al., 1995; Moga et al., 1995; Bubser and Deutch, 1998; Otake and Nakamura, 1998; Pinto et al., 2003), mainly targeting inner layers (5/6) of IL (Be- rendse and Groenewegen, 1991; Pinto et al., 2003). We recently identified labeled cells rostrocaudally throughout PV following retrograde tracer injections in IL (Hoover and Vertes, 2007). PV also distributes substantially to area 25 (or the infralimbic cortex) in primates (Hsu and Price, 2007). Moga et al. (1995) reported that PVa projects densely, whereas PVp sparsely (or not at all) to the lateral septum (LS). We showed that both PVa and PVp project to LS, but similar to Moga et al. (1995) found that the major output was from PVa. Consistent with this, Risold and Swanson (1997) described labeled cells throughout PV following FG injections in LS, but progressively fewer cells at successive caudal levels of PV. Fig. 11. A–C: Series of rostrocaudally aligned low magnification Few reports have examined PV projections to the bed darkfield photomicrographs of transverse sections through the forebrain depicting patterns of labeling within the ventral subiculum nucleus of stria terminalis (Moga et al., 1995; Van der (SUBv) and lateral entorhinal cortex (ECl) produced by a PHA-L Werf et al., 2002). In general accord with present findings, injection into the paratenial nucleus of the thalamus. Note strong Moga et al. (1995) reported that PV distributes heavily to labeling in ECl as well as in the molecular layer of SUBv. Scale bar ϭ rostral and lateral parts (subnuclei) of BST. Although the 500 ␮m. See list for abbreviations. efferent projections of BST have been fairly extensively examined (Dong et al., 2001; Gu et al., 2003; Dong and Swanson, 2004, 2006), to our knowledge, only a single PV projections: comparisons with previous early report by Weller and Smith (1982) examined afferents to BST. They showed PV and PT are virtually the sole studies sources of thalamic input to BST, distributing signifi- As discussed, the output of PV is restricted; that is, PV cantly to BST. projects to a limited number of sites, but quite massively We showed that PV distributes massively throughout to them. The foremost PV targets are nucleus accumbens, the amygdala, and with the exception of parts of the bed nucleus of stria terminalis, and the amygdala. caudal amygdala, to most subnuclei of the amygdala. The The most complete analysis of PV projections was an foremost PV targets are the central and basal nuclei of the early report by Moga et al. (1995). Our findings were amygdala. In accord with present findings, an early exam- comparable to theirs for the anterior PV (PVa) but differed ination of thalamic afferents to the amygdala (Ottersen The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 231

Fig. 12. A–C: Series of low-magnification brightfield photomicrographs of transverse sections through the forebrain depicting the site graphs of transverse sections through the forebrain depicting the site of a FluoroGold injection in the core of nucleus accumbens (ACCs) (D) of a FluoroGold injection in the shell of nucleus accumbens (ACCs) (A) and patterns of retrogradely labeled cells within the anterior para- and patterns of retrogradely labeled cells within the anterior paraventricular (PV) and paratenial (PT) nuclei (E) and posterior paraventricular (PV) and paratenial (PT) nuclei (B) and the posterior ventricular nucleus (F) produced by this injection. Note significant paraventricular nucleus (C) produced by this injection. Note signifi- number of retrogradely labeled neurons in anterior PV and PT and cant numbers of retrogradely labeled neurons in PT, moderate num- moderately number in posterior PV produced by this injection. Scale bers in posterior PV, and relatively few in the anterior PV with this bar ϭ 500 ␮m for A,D; 350 ␮m for B,E; 400 ␮m for C,F. See list for injection. D–F: Series of low- magnification brightfield photomicro- abbreviations. The Journal of Comparative Neurology

232 R.P. VERTES AND W.B. HOOVER and Ben-Ari, 1979) described widespread PV (and PT) ACC overlap with zones of strong and weak enkephalin projections to the amygdala, stating that “the paraven- immunoreactivity, respectively. We did not immunostain tricular and paratenial nuclei of the thalamus were found for enkephalin and, hence, cannot confirm these findings. to project throughout the amygdaloid complex.” Several We showed that PV distributes moderately to the ento- subsequent studies have confirmed pronounced PV projec- rhinal cortex and to the ventral subiculum of the tions to CEA, BMA, and BLA (Berendse and Groenewe- hippocampus—mainly to the rostral EC/subiculum and to gen, 1990; Su and Bentivoglio, 1990; Turner and Herken- ventral aspects of the subiculum, adjoining EC. Previous ham, 1991; Moga et al., 1995; Peng and Bentivoglio, 2004). reports (Berendse and Groenewegen, 1991; Moga et al., PV projections to ‘cognitive-associated’ forebrain 1995) have similarly described PV projections to EC and to areas. Groenewegen and colleagues (Room et al., 1985; the ventral subiculum and, like here, stronger projections Groenewegen et al., 1990) initially defined a system of from the anterior than posterior PV. Retrograde tracer connections (or loop) from the prelimbic cortex Ͼ ventral injections in the hippocampus, involving the ventral sub- striatum Ͼ ventral pallidum Ͼ MD of thalamus Ͼ PL iculum, give rise to labeled cells in PV (mainly PVa) (Wyss which they termed the ‘PL circuit.’ The ‘PL circuit’ has et al., 1979; Riley and Moore, 1981; Su and Bentivoglio, subsequently been expanded to include several additional 1990), and the hippocampus is the source of significant structures; principal among them are the dorsal agranu- return projections to PV, originating from the ventral lar insular cortex (AId), hippocampus/parahippocampus, subiculum (Witter, 2006). the basolateral amygdala, parts of midline thalamus and As described, the amygdala is a major PV target, with the ventral tegmental area (VTA) (Vertes, 2006). PL and projections heaviest to the central (CEA) and basomedial its interconnected circuitry serve a recognized role in cog- (BMA) nuclei of amygdala. While PV projections to BLA nitive functions (Laroche et al., 2000; Groenewegen and are less dense than to CEA and BMA, they are nonethe- Uylings, 2000; Vertes, 2006). PV distributes to several less pronounced, mainly targeting the posterior BLA structures of prelimbic circuit: PL, AId, ACC, EC, the (BLAp), bordering BMA. Earlier reports have similarly ventral subiculum, BLA, and VTA. described marked PV projections to BLA (Ottersen and We showed that PV (PVa and PVp) distributes: 1) se- Ben-Ari, 1979; Berendse and Groenewegen, 1990, 1991; lectively to IL and PL of the ventral mPFC; 2) more Su and Bentivolglio, 1990; Turner and Herkenham, 1991; heavily to PL than to IL; and 3) rostrocaudally throughout Moga et al., 1995). BLA is an integral part of the “prelim- PL, terminating most densely in inner layers of PL. These bic circuit.” BLA has strong links with PL (Sesack et al., findings are consistent with previous descriptions of sig- 1989; McDonald, 1987, 1991; McDonald et al., 1996; nificant PV projections to PL (Berendse and Groenewegen, Conde et al., 1995; Vertes, 2004; Gabbott et al., 2006; 1991; Conde et al., 1995; Moga et al., 1995; Bubser and Hoover and Vertes, 2007), as well as with other parts of Deutch, 1998; Otake and Nakamura, 1998; Pinto et al., the circuit including the hippocampus, ACC, claustrum, 2003; Hoover and Vertes, 2007). and the insular cortex (McDonald, 1987; Brog et al., 1993; We found that PVa and PVp distribute massively Petrovich et al., 1996; Pikkarainen et al., 1999; Majak et throughout the shell and core of ACC. Although early al., 2002). reports showed that PV (and PT) strongly target ACC PV as an interface in the flow of information between (Groenewegen et al., 1980; Newman and Winans, 1980; the suprachiasmatic nucleus (SCN) and other regions Beckstead, 1984; Jayaraman, 1985; Phillipson and Grif- fiths, 1985), Groenewegen and colleagues (Berendse et al., of the brain. In accord with previous reports (Moga et 1988; Berendse and Groenewegen, 1990) were the first to al., 1995; Moga and Moore, 1997; Krout et al., 2002), we show that each of the midline nuclei distribute to select, showed that PV projects moderately densely to the supra- and only partially overlapping, territories of the ventral chiasmatic nucleus of the hypothalamus. SCN, in turn, is striatum. Regarding PV, they reported that of the midline the source of significant projections to PV (Watts et al., nuclei of thalamus, PV was the predominant source of 1987; Novak et al., 2000; Peng and Bentivoglio, 2004; afferents to the shell of ACC, and while also pronounced to Zhang et al., 2006). Accordingly, PV appears to represent the core, were shared by other midline thalamic groups to an important relay in the transfer of information to and the core (Berendse and Groenewegen, 1990). Several sub- from the SCN—the circadian pacemaker (Mistlberger, sequent studies have confirmed ‘massive’ PV projections 2005; Morin and Allen, 2006). While afferents to SCN to ACC, and further showed that a fairly significant per- generally serve to entrain SCN activity to light/dark con- centage of PV fibers to ACC collateralize to other sites ditions, PV lesions do not disrupt circadian timing or (Meredith and Wouterlood, 1990; Su and Bentivoglio, entrainment to light (Ebling et al., 1992). This suggests a 1990; Brog et al., 1993; Freedman and Cassell, 1994; Moga nonphotic modulatory influence of PV on SCN. Moga et al. et al., 1995; Bubser and Deutch, 1998; Otake and Naka- (1995) proposed that PV conveys information on basal mura, 1998; Erro et al., 2002; Pinto et al., 2003; Parsons et levels of activation to the SCN—functions associated with al., 2006, 2007), mainly to the mPFC (Bubser and Deutch, PV/midline thalamus (Van der Werf et al., 2002; Vertes, 1998; Otake and Nakamura, 1998) and to the amygdala 2006). (Su and Bentivoglio, 1990). Regarding SCN-PV projections, the SCN has few direct Finally, in accord with previous reports (Berendse et al., outputs to the systems it affects (Deurveilher and Semba, 1988; Berendse and Groenewegen, 1990), we found that 2005; Morin and Allen, 2006), indicating indirect routes to PV fibers distribute in a nonhomogeneous (or ‘patch/ them, possibly through PV. At the light and EM levels, matrix’) manner to the nucleus accumbens; that is, re- Peng and Bentivoglio (2004) showed that SCN strongly gions of dense innervation interspersed with relatively targets PV, and further that SCN fibers synapse with PV fiber free zones (Figs. 2, 3B, 4, 5A). Berendse et al. (1988) cells projecting to the amygdala. On this basis they con- reported that densely PV-innervated regions of the rostral cluded that PV “plays a role in the transfer of circadian ACC and sparsely innervated areas of the caudomedial timing information to the limbic system.” The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 233

Figure 13 The Journal of Comparative Neurology

234 R.P. VERTES AND W.B. HOOVER

PT projections: comparisons with previous CEA, projecting instead to the fringes of CEA. Ottersen studies and Ben-Ari (1979) identified few labeled cells in PT with retrograde injections in CEA, the cortical nuclei, or ante- As discussed, there is significant overlap in PV and PT rior regions of the basal nuclei, but described significant projections. Although a few reports have described PT numbers of reacted cells in PT following large amygdalar projections to specific targets, to our knowledge no previ- injections spanning the rostral and caudal BMA/BLA. ous study has examined the totality of PT projections. Subsequent reports using retrograde (Su and Bentivoglio, We showed that PT distributes densely throughout the 1990) or anterograde tracers (Turner and Herkenham, ventral PFC to AC, PL, IL, and medial orbital (MO) cor- 1991) similarly demonstrated PT projections to the lat- tices, with projections heaviest to PL. In an early exami- eral, basomedial, and basolateral nuclei of amygdala. nation of midline and intralaminar thalamic connections We showed that PT distributes throughout entorhinal with the cortex, Berendse and Groenewegen (1991) simi- cortex and the ventral subiculum, terminating within larly described PT projections to the mPFC, but unlike fairly restricted zones of both sites: mainly inner layers here, projections were modest and largely confined to ven- (3–6) of the lateral EC and the molecular layer of the tral aspects of the mPFC, mainly to MO and IL. Their PT ventral subiculum. Berendse and Groenewegen (1991) injection, however, was small and restricted to the medial demonstrated a similar distribution of PT fibers to EC and aspect of the anterior PT (see their fig. 1A, p. 75; Berendse to the ventral subiculum, but in contrast to the present and Groenewegen, 1991). Supporting present findings, findings described stronger PV than PT projections to retrograde tracer injections in AC, PL, and IL have been these sites—we found the opposite. Differences could in- shown to give rise to significant numbers of labeled cells in volve relative size and placements of injections in PT and PT (Conde et al., 1995; Hoover and Vertes, 2007). PT also PV. Several reports have identified labeled cells in PT strongly targets the ventral mPFC in primates, mainly IL following retrograde tracer injections in the hippocampus (area 25) and PL (area 32) (Hsu and Price, 2007). (Wyss et al., 1979; Riley and Moore, 1981; Su and Ben- Similar to PV, ACC is a major destination of PT fibers. tivoglio, 1990), or entorhinal cortex (Beckstead, 1978; In general accord with present results, two early studies Wyss et al., 1979; Insausti et al., 1987). (Kelley and Stinus, 1984; Carlsen and Heimer, 1986) described robust PT projections to ACC, terminating heavily Functional considerations in the medial two-thirds of ACC (shell region) with some Although the projections of PV and PT significantly extension dorsally into dorsomedial aspects of CP. Be- overlap, suggesting comparable functions, considerably rendse and Groenewegen (1990) confirmed marked PT greater attention has been given to the functional charac- projections to the shell of ACC, particularly dense to the teristics of PV. An accumulating body of evidence indi- ventromedial shell of ACC. Although they also reported cates that PV receives inputs from several sites of the that PT distributes to the core of ACC and to the dorso- brainstem and hypothalamus that are known to exert medial CP, projections were considerably less pronounced activating and/or ‘wakefulness-promoting’ effects on the than the presently described massive distribution of PT forebrain. This includes afferents from monoaminergic, fibers to the core of ACC and to the dorsomedial CP (see cholinergic, and peptide-containing systems of the brain- Figs. 7D–F, 10A). Consistent with our results, Brog et al. stem and diencephalon, prominently including orexin/ (1993) described significant numbers of labeled cells in PT hypocretin cells of the lateral hypothalamus (Chen and following retrograde tracer injections in the core or shell of Su, 1990; Vertes, 1991; Freedman and Cassell, 1994; ACC. Otake and Ruggiero, 1995; Peyron et al., 1998; Cutler et As demonstrated, PT fibers mainly target caudal re- al., 1999; Vertes et al., 1999; Bhatnagar et al., 2000; Krout gions of the amygdala, predominantly the lateral and et al., 2002; Kirouac et al., 2005; Otake, 2005; Parsons et basal nuclei of amygdala. In contrast to the dense PV al., 2006). Accordingly, PV/PT (and other nuclei of the innervation of CEA, PT fibers largely avoid the core of midline thalamus) are thought to serve an essential role in arousal and attention (Van der Werf et al., 2002; Vertes, 2006; Vertes et al., 2006). In line with the foregoing, PV cells show elevated levels Fig. 13. A–C: Series of low-magnification brightfield photomicro- of c-fos expression during wakefulness (Peng et al., 1995; graphs of transverse sections through the forebrain depicting the site Novak et al., 2000) as well as during stressful conditions, of a FluoroGold injection in the basolateral nucleus (BLA) of the elicited by various stressors (Chastrette et al., 1991; Bub- amygdala (A) and patterns of retrogradely labeled cells within the anterior paraventricular (PV) and paratenial (PT) nuclei (B) and the ser and Deutch, 1999; Sica et al., 2000; Otake et al., 2002). posterior paraventricular nucleus (C) produced by this injection. Note PV appears to be critically involved in adaptive responses significant numbers of retrogradely labeled neurons in the posterior to stress (Bhatnagar and Dallman, 1998; Bhatnagar et al., PV but relatively few numbers in the anterior PV and PT with this 2000, 2002; Otake et al., 2002) through direct (Sawchenko injection. D–F: Series of low-magnification brightfield photomicro- and Swanson, 1983; present results), or predominantly graphs of transverse sections through the forebrain depicting the site indirect projections to paraventricular nucleus of the hy- of a FluoroGold injection in the central nucleus (CEA) of the amygdala (D) and patterns of retrogradely labeled cells within the anterior pothalamus (Sawchenko and Swanson, 1983; Bhatnagar paraventricular (PV) and paratenial (PT) nuclei (E) and the posterior and Dallman, 1998; Dong et al., 2001; Otake et al., 2002; paraventricular nucleus (F) produced by this injection. Note moderate Dong and Swanson, 2006). numbers of retrogradely labeled cells in anterior and posterior PV, In addition to its role in promoting arousal/wakefulness, and relatively few in PT. Finally, note the presence of labeled neurons the orexin system participates in feeding behavior (for in other nuclei of the midline thalamus (C,F) produced with BLA and review, see Willie et al., 2001). Intraventricular injections CEA injections; namely, in the intermediodorsal and central medial nuclei (C) and the rhomboid and reuniens nuclei (F). Scale bar ϭ 750 of orexin (orexin A) stimulates food consumption in sati- ␮m for A; 300 ␮m for B; 500 ␮m for C; 700 ␮m for D; 400 ␮m for E; 450 ated rats (Sakurai et al., 1998; Edwards et al., 1999; ␮m for F. See list for abbreviations. Haynes et al., 1999), anti-orexin antibodies or receptor The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 235 antagonists suppress feeding (Haynes et al., 2000), and prefrontal corticostriatal projections in the rat. J Comp Neurol 316: orexin knockout mice are hypophagic (Hara et al., 2001). 314–447. These effects may in part be mediated the actions of orexin Bertram EH, Zhang DX. 1999. Thalamic excitation of hippocampal CA1 neurons: a comparison with the effects of CA3 stimulation. Neuro- on PV. Specifically, Nakahara et al. (2004) described ele- science 92:15–26. vated c-fos expression in PV in anticipation of feeding Bhatnagar S, Dallman M. 1998. Neuroanatomical basis for facilitation of (food anticipatory activity) in deprived rats, and further hypothalamic-pituitary-adrenal responses to a novel stressor after reported that PV lesions attenuated anticipatory locomo- chronic stress. Neuroscience 84:1025–1039. tor activity associated with feeding. Angeles-Castellanos Bhatnagar S, Viau V, Chu A, Soriano L, Meijer OC, Dallman MF. 2000. A et al. (2007) similarly reported elevated c-fos expression in cholecystokinin-mediated pathway to the paraventricular thalamus is PV with anticipated feeding and, interestingly, also de- recruited in chronically stressed rats and regulates hypothalamic- scribed enhanced c-fos expression in several other limbic pituitary-adrenal function. J Neurosci 20:5564–5573. Bhatnagar S, Huber R, Nowak N, Trotter P. 2002. Lesions of the posterior forebrain structures before and immediately after the de- paraventricular thalamus block habituation of hypothalamic-pituitary- livery of food to deprived rats, including the central and adrenal responses to repeated restraint. J Neuroendocrinol 14:403– basal nuclei of the amygdala, BST, the lateral septum, 410. ACC (core and shell) and the infralimbic/prelimbic corti- Brog JS, Salyapongse A, Deutch AY, Zahm DS. 1993. The patterns of ces. As demonstrated here, these are the main forebrain afferent innervation of the core and shell in the “accumbens” part of the targets of PV (and PT), suggesting a PV/PT influence on rat ventral striatum: immunohistochemical detection of retrogradely them in complex behaviors associated with feeding. transported fluoro-gold. J Comp Neurol 338:255–278. Bubser M, Deutch AY. 1998. Thalamic paraventricular nucleus neurons Related to this, Kelley et al. (2005) recently put forth a collateralize to innervate the prefrontal cortex and nucleus accumbens. model suggestive of a role for PV in motivated behaviors Brain Res 787:304–310. as exemplified by feeding. Although not necessarily lim- Bubser M, Deutch AY. 1999. Stress induces Fos expression in neurons of ited to feeding, the model addressed mechanisms respon- the thalamic paraventricular nucleus that innervate limbic forebrain sible for the ingestion of palatable foods under satiated sites. Synapse 32:13–22. conditions. According to Kelley et al. (2005), PV receives Carlsen J, Heimer L. 1986. The projection from the parataenial thalamic diverse inputs which, among other things, code the incen- nucleus, as demonstrated by the Phaseolus vulgaris-leucoagglutinin (PHA-L) method, identifies a subterritorial organization of the ventral tive value of foods and when incentives are high (desirable striatum. Brain Res 374:375–379. foods), feeding ensues, even in a satiated state, mainly Chastrette N, Pfaff DW, Gibbs RB. 1991. Effects of daytime and nighttime through the actions of PV on the ventral striatum. They stress on Fos-like immunoreactivity in the paraventricular nucleus of stated: “PVT may act as an interface between signals the hypothalamus, the habenula, and the posterior paraventricular related to arousal, energy balance, circadian or diurnal nucleus of the thalamus. Brain Res 563:339–344. rhythms, and reward, and major striatal motor output Chen S, Su H-S. 1990. Afferent connections of the thalamic paraventricular systems.” and parataenial nuclei in the rat—a retrograde tracing study with iontophoretic application of Fluoro-Gold. Brain Res 522:1–6. In summary, PV (and PT) receive a vast array of affer- Conde F, Maire-Lepoivre E, Audinat E, Crepel F. 1995. Afferent connec- ents from the brainstem, hypothalamus, and limbic fore- tions of the medial frontal cortex of the rat. II. Cortical and subcortical brain and appear to serve as a critical gateway for the afferents. J Comp Neurol 352:567–593. transfer of multimodal information to structures of the Cutler DJ, Morris R, Sheridhar V, Wattam TA, Holmes S, Patel S, Arch JR, limbic system in the selection of appropriate responses to Wilson S, Buckingham RE, Evans ML, Leslie RA, Williams G. 1999. changing environmental conditions. In effect, depending Differential distribution of orexin-A and orexin-B immunoreactivity in on the relative dominance of sets of inputs to PV/PT, the rat brain and spinal cord. Peptides 20:1455–1470. coupled with their output to specific structures of the Dempsey EW, Morison RS. 1942. The production of rhythmically recurrent cortical potentials after localized thalamic stimulation. Am J Physiol limbic forebrain, PV/PT would guide behavior toward a 135:293–300. particular outcome among various potential outcomes. Deurveilher S, Semba K. 2005. Indirect projections from the suprachiasmatic nucleus to major arousal-promoting cell groups in rat: implications for the circadian control of behavioural state. Neuroscience 130: LITERATURE CITED 165–183. Angeles-Castellanos M, Mendoza J, Escobar C. 2007. Restricted feeding Dolleman-Van der Weel MJ, Lopes da Silva FH, Witter MP. 1997. Nucleus schedules phase shift daily rhythms of c-Fos and protein Per1 immu- reuniens thalami modulates activity in hippocampal field CA1 through noreactivity in corticolimbic regions in rats. Neuroscience 144:344– excitatory and inhibitory mechanisms. J Neurosci 17:5640–5650. 355. Dong HW, Swanson LW. 2004. Projections from bed nuclei of the stria Beckstead RM. 1978. Afferent connections of the entorhinal area in the rat terminalis, posterior division: implications for cerebral hemisphere as demonstrated by retrograde cell-labeling with horseradish peroxi- regulation of defensive and reproductive behaviors. J Comp Neurol dase. Brain Res 152:249–264. 471:396–433. Beckstead RM. 1984. The thalamostriatal projection in the cat. J Comp Dong HW, Swanson LW. 2006. Projections from bed nuclei of the stria Neurol 223:313–346. terminalis, anteromedial area: cerebral hemisphere integration of neu- Bentivoglio M, Balercia G, Kruger L. 1991. The specificity of the nonspe- roendocrine, autonomic, and behavioral aspects of energy balance. cific thalamus: the midline nuclei. Prog Brain Res 87:53–80. J Comp Neurol 494:142–178. Berendse HW, Groenewegen HJ. 1990. Organization of the thalamostriatal Dong HW, Petrovich GD, Watts AG, Swanson LW. 2001. Basic organiza- projections in the rat, with special emphasis on the ventral striatum. tion of projections from the oval and fusiform nuclei of the bed nuclei of J Comp Neurol 299:187–228. the stria terminalis in adult rat brain. J Comp Neurol 436:430–455. Berendse HW, Groenewegen HJ. 1991. Restricted cortical termination Ebling FJ, Maywood ES, Humby T, Hastings MH. 1992. Circadian and fields of the midline and intralaminar thalamic nuclei in the rat. photoperiodic time measurement in male Syrian hamsters following Neuroscience 42:73–102. lesions of the melatonin-binding sites of the paraventricular thalamus. Berendse HW, Voorn P, te Kortschot A, Groenewegen HJ. 1988. Nuclear J Biol Rhythms 7:241–254. origin of thalamic afferents of the ventral striatum determines their Edwards CM, Abusnana S, Sunter D, Murphy KG, Ghatei MA, Bloom SR. relation to patch/matrix configurations in enkephalin-immuno- 1999. The effect of the orexins on food intake: comparison with neu- reactivity in the rat. J Chem Neuroanat 1:3–10. ropeptide Y, melanin-concentrating hormone and galanin. J Endocrinol Berendse HW, Galis-de Graaf Y, Groenewegen HJ. 1992. Topographical 160:R7–12. organization and relationship with ventral striatal compartments of Erro ME, Lanciego JL, Gimenez-Amaya JM. 2002. Re-examination of the The Journal of Comparative Neurology

236 R.P. VERTES AND W.B. HOOVER

thalamostriatal projections in the rat with retrograde tracers. Neurosci McDonald AJ. 1987. Organization of amygdaloid projections to the me- Res 42:45–55. diodorsal thalamus and prefrontal cortex: a fluorescence retrograde Freedman LJ, Cassell MD. 1994. Relationship of thalamic basal forebrain transport study in the rat. J Comp Neurol 262:46–58. projection neurons to the peptidergic innervation of the midline thal- McDonald AJ. 1991. Organization of amygdaloid projections to the prefron- amus. J Comp Neurol 348:321–342. tal cortex and associated striatum in the rat. Neuroscience 44:1–14. Gabbott PL, Warner TA, Busby SJ. 2006. Amygdala input monosynapti- McDonald AJ, Mascagni F, Guo L. 1996. Projections of the medial and cally innervates parvalbumin immunoreactive local circuit neurons in lateral prefrontal cortices to the amygdala: a Phaseolus vulgaris leu- rat medial prefrontal cortex. Neuroscience 139:1039–1048. coagglutinin study in the rat. Neuroscience 71:55–75. Groenewegen HJ, Berendse HW. 1994. The specificity of the “non-specific” Meredith GE, Wouterlood FG. 1990. Hippocampal and midline thalamic midline and intralaminar thalamic nuclei. Trends Neurosci 17:52–57. fibers and terminals in relation to the choline acetyltransferase- Groenewegen HJ, Uylings HB. 2000. The prefrontal cortex and the inte- immunoreactive neurons in nucleus accumbens of the rat: a light and gration of sensory, limbic and autonomic information. Prog Brain Res electron microscopic study. J Comp Neurol 296:204–221. 126:3–28. Mistlberger RE. 2005. Circadian regulation of sleep in mammals: role of Groenewegen HJ, Witter MP. 2004. Thalamus. In: Paxinos G, editor. The the suprachiasmatic nucleus. Brain Res Brain Res Rev 49:429–454. rat nervous system, 3rd ed. New York: Academic Press. p 408–453. Moga MM, Moore RY. 1997. Organization of neural inputs to the supra- Groenewegen HJ, Becker NE, Lohman AH. 1980. Subcortical afferents of chiasmatic nucleus in the rat. J Comp Neurol 389:508–534. the nucleus accumbens septi in the cat, studied with retrograde axonal Moga MM, Weis RP, Moore RY. 1995. Efferent projections of the paraven- transport of horseradish peroxidase and bisbenzimid. Neuroscience tricular thalamic nucleus in the rat. J Comp Neurol 359:221–238. 5:1903–1916. Morin LP, Allen CN. 2006. The circadian visual system. Brain Res Rev Groenewegen HJ, Berendse HW, Wolters JG, Lohman AH. 1990. The 51:1–60. anatomical relationship of the prefrontal cortex with the striatopallidal Nakahara K, Fukui K, Murakami N. 2004. Involvement of thalamic para- system, the thalamus and the amygdala: evidence for a parallel orga- ventricular nucleus in the anticipatory reaction under food restriction nization. Prog Brain Res 85:95–116. in the rat. J Vet Med Sci 66:1297–1300. Groenewegen HJ, Galis-de Graaf Y, Smeets WJAJ. 1999. Integration and Newman R, Winans SS. 1980. An experimental study of the ventral stri- segregation of limbic cortico-striatal loops at the thalamic level: an atum of the golden hamster. I. Neuronal connections of the nucleus experimental tracing study in rats. J Chem Neuroanat 16:167–185. accumbens. J Comp Neurol 191:167–192. Gu G, Cornea A, Simerly RB. 2003. Sexual differentiation of projections Novak CM, Harris JA, Smale L, Nunez AA. 2000. Suprachiasmatic nucleus from the principal nucleus of the bed nuclei of the stria terminalis. projections to the paraventricular thalamic nucleus in nocturnal rats J Comp Neurol 460:542–562. (Rattus norvegicus) and diurnal nile grass rats (Arviacanthis niloti- Hara J, Beuckmann CT, Nambu T, Willie JT, Chemelli RM, Sinton CM, cus). Brain Res 874:147–157. Sugiyama F, Yagami K, Goto K, Yanagisawa M, Sakurai T. 2001. Otake K. 2005. Cholecystokinin and substance P immunoreactive projec- Genetic ablation of orexin neurons in mice results in narcolepsy, hy- tions to the paraventricular thalamic nucleus in the rat. Neurosci Res pophagia, and obesity. Neuron 30:345–354. 51:383–394. Haynes AC, Jackson B, Overend P, Buckingham RE, Wilson S, Tadayyon Otake K, Nakamura Y. 1998. Single midline thalamic neurons projecting M, Arch JR. 1999. Effects of single and chronic intracerebroventricular to both the ventral striatum and the prefrontal cortex in the rat. administration of the orexins on feeding in the rat. Peptides 20:1099– Neuroscience 86:635–649. 1105. Otake K, Ruggiero DA. 1995. Monoamines and nitric oxide are employed by Haynes AC, Jackson B, Chapman H, Tadayyon M, Johns A, Porter RA, afferents engaged in midline thalamic regulation. J Neurosci 15:1891– Arch JR. 2000. A selective orexin-1 receptor antagonist reduces food 1911. consumption in male and female rats. Regul Pept 96:45–51. Otake K, Kin K, Nakamura Y. 2002. Fos expression in afferents to the rat Hoover WB, Vertes RP. 2007. Anatomical analysis of afferent projections to midline thalamus following immobilization stress. Neurosci Res 43: the medial prefrontal cortex in the rat. Brain Struc Funct 212:149–179. 269–282. Hsu DT, Price JL. 2007. Midline and intralaminar thalamic connections Ottersen OP, Ben-Ari Y. 1979. Afferent connections to the amygdaloid with the orbital and medial prefrontal networks in macaque monkeys. complex of the rat and cat. I. Projections from the thalamus. J Comp J Comp Neurol 504:89–111. Neurol 187:401–424. Insausti R, Amaral DG, Cowan WM. 1987. The entorhinal cortex of the Parsons MP, Li S, Kirouac GJ. 2006. The paraventricular nucleus of the monkey: III. Subcortical afferents. J Comp Neurol 264:396–408. thalamus as an interface between the orexin and CART peptides and Jayaraman A. 1985. Organization of thalamic projections in the nucleus the shell of the nucleus accumbens. Synapse 59:480–490. accumbens and the caudate nucleus in cats and its relation with hip- Parsons MP, Li S, Kirouac GJ. 2007. Functional and anatomical connection pocampal and other subcortical afferents. J Comp Neurol 231:396–420. between the paraventricular nucleus of the thalamus and dopamine Kelley AE, Stinus L. 1984. The distribution of the projection from the fibers of the nucleus accumbens. J Comp Neurol 500:1050–1063. parataenial nucleus of the thalamus to the nucleus accumbens in the Peng ZC, Bentivoglio M. 2004. The thalamic paraventricular nucleus re- rat: an autoradiographic study. Exp Brain Res 54:499–512. lays information from the suprachiasmatic nucleus to the amygdala: a Kelley AE, Baldo BA, Pratt WE. 2005. A proposed hypothalamic-thalamic- combined anterograde and retrograde tracing study in the rat at the striatal axis for the integration of energy balance, arousal, and food light and electron microscopic levels. J Neurocytol 33:101–116. reward. J Comp Neurol 493:72–85. Peng ZC, Grassi-Zucconi G, Bentivoglio M. 1995. Fos-related protein ex- Kirouac GJ, Parsons MP, Li S. 2005. Orexin (hypocretin) innervation of the pression in the midline paraventricular nucleus of the rat thalamus: paraventricular nucleus of the thalamus. Brain Res 1059:179–188. basal oscillation and relationship with limbic efferents. Exp Brain Res Kirouac GJ, Parsons MP, Li S. 2006. Innervation of the paraventricular 104:21–29. nucleus of the thalamus from cocaine- and amphetamine-regulated Petrovich GD, Risold PY, Swanson LW. 1996. Organization of projections transcript (CART) containing neurons of the hypothalamus. J Comp from the basomedial nucleus of the amygdala: a PHAL study in the rat. Neurol 497:155–165. J Comp Neurol 374:387–420. Krout KE, Belzer RE, Loewy AD. 2002. Brainstem projections to midline Peyron C, Tighe DK, van den Pol AN, de Lecea L, Heller HC, Sutcliffe JG, and intralaminar thalamic nuclei of the rat. J Comp Neurol 448:53– Kilduff TS. 1998. Neurons containing hypocretin (orexin) project to 101. multiple neuronal systems. J Neurosci 18:9996–10015. Kung JC, Shyu BC. 2002. Potentiation of local field potentials in the Phillipson OT, Griffiths AC. 1985. The topographic order of inputs to anterior cingulate cortex evoked by the stimulation of the medial tha- nucleus accumbens in the rat. Neuroscience 16:275–296. lamic nuclei in rats. Brain Res 953:37–44. Pikkarainen M, Ronkko S, Savander V, Insausti R, Pitkanen A. 1999. Laroche S, Davis S, Jay TM. 2000. Plasticity at hippocampal to prefrontal Projections from the lateral, basal, and accessory basal nuclei of the cortex synapses: dual roles in working memory and consolidation. amygdala to the hippocampal formation in rat. J Comp Neurol 403: Hippocampus 10:438–446. 229–260. Majak K, Pikkarainen M, Kemppainen S, Jolkkonen E, Pitkanen A. 2002. Pinto A, Jankowski M, Sesack SR. 2003. Projections from the paraven- Projections from the amygdaloid complex to the claustrum and the tricular nucleus of the thalamus to the rat prefrontal cortex and nu- endopiriform nucleus: a Phaseolus vulgaris leucoagglutinin study in cleus accumbens shell: ultrastructural characteristics and spatial re- the rat. J Comp Neurol 451:236–249. lationships with dopamine afferents. J Comp Neurol 459:142–155. The Journal of Comparative Neurology

EFFERENTS OF PV AND PT NUCLEI 237

Riley JN, Moore RY. 1981. Diencephalic and brainstem afferents to the Vertes RP. 2002. Analysis of projections from the medial prefrontal cortex hippocampal formation of the rat. Brain Res Bull 6:437–444. to the thalamus in the rat, with emphasis on nucleus reuniens. J Comp Risold PY, Swanson LW. 1997. Connections of the rat lateral septal com- Neurol 442:163–187. plex. Brain Res Rev 24:115–195. Vertes RP. 2004. Differential projections of the infralimbic and prelimbic Room P, Russchen FT, Groenewegen HJ, Lohman AHM. 1985. Efferent cortex in the rat. Synapse 51:32–58. connections of the paralimbic (area 32) and the infralimbic (area 25) Vertes RP. 2006. Interactions among the medial prefrontal cortex, hip- cortices: an anterograde tracing study in the cat. J Comp Neurol pocampus and midline thalamus in emotional and cognitive processing 242:40–55. in the rat. Neuroscience 142:1–20. Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H, Vertes RP, Fortin WJ, Crane AM. 1999. Projections of the median raphe Williams SC, Richardson JA, Kozlowski GP, Wilson S, Arch JR, Buck- nucleus in the rat. J Comp Neurol 407:555–582. ingham RE, Haynes AC, Carr SA, Annan RS, McNulty DE, Liu WS, Terrett JA, Elshourbagy NA, Bergsma DJ, Yanagisawa M. 1998. Orex- Vertes RP, Hoover WB, Do Valle AC, Sherman A, Rodriguez JJ. 2006. ins and orexin receptors: a family of hypothalamic neuropeptides and G Efferent projections of reuniens and rhomboid nuclei of the thalamus in protein-coupled receptors that regulate feeding behavior. Cell 92:573– the rat. J Comp Neurol 499:768–796. 585. Viana Di Prisco G, Vertes RP. 2006. Excitatory actions of the ventral Sawchenko PE, Swanson LW. 1983. The organization and biochemical midline thalamus (rhomboid/reuniens) on the medial prefrontal cortex speciﬁcity of afferent projections to the paraventricular and supraoptic in the rat. Synapse 60:45–55. nuclei. Prog Brain Res 60:19–29. Watts AG, Swanson LW, Sanchez-Watts G. 1987. Efferent projections of Sesack SR, Deutch AY, Roth RH, Bunney BS. 1989. Topographical orga- the suprachiasmatic nucleus: I. Studies using anterograde transport of nization of the efferent projections of the medial prefrontal cortex in the Phaseolus vulgaris leucoagglutinin in the rat. J Comp Neurol 258:204– rat: an anterograde tract-tracing study with Phaseolus vulgaris leuco- 229. agglutinin. J Comp Neurol 290:213–242. Weller KL, Smith DA. 1982. Afferent connections to the bed nucleus of the Sica AL, Greenberg HE, Scharf SM, Ruggiero DA. 2000. Chronic- stria terminalis. Brain Res 232:255–270. intermittent hypoxia induces immediate early gene expression in the midline thalamus and epithalamus. Brain Res. 883:224–228. Willie JT, Chemelli RM, Sinton CM, Yangisawa M. 2001. To eat or sleep: orexin in the regulation of feeding and wakefulness. Annu Rev Neuro- Su HS, Bentivoglio M. 1990. Thalamic midline cell populations projecting sci 24:429–458. to the nucleus accumbens, amygdala, and hippocampus in the rat. J Comp Neurol 297:582–593. Witter MP. 2006. Connections of the subiculum of the rat: topography in relation to columnar and laminar organization. Behav Brain Res 174: Swanson LW. 1998. Brain maps: structure of the rat brain. New York: Elsevier. 251–264. Turner BH, Herkenham M. 1991. Thalamoamygdaloid projections in the Wyss JM, Swanson LW, Cowan WM. 1979. A study of subcortical afferents rat: a test of the amygdala’s role in sensory processing. J Comp Neurol to the hippocampal formation in the rat. Neuroscience 4:463–476. 313:295–325. Zhang DX, Bertram EH. 2002. Midline thalamic region: widespread exci- Van der Werf YD, Witter MP, Groenewegen HJ. 2002. The intralaminar tatory input to the entorhinal cortex and amygdala. J Neurosci 22: and midline nuclei of the thalamus. Anatomical and functional evi- 3277–3284. dence for participation in processes of arousal and awareness. Brain Zhang L, Kolaj M, Renaud LP. 2006. Suprachiasmatic nucleus communi- Res Rev 39:107–140. cates with anterior thalamic paraventricular nucleus neurons via rapid Vertes RP. 1991. A PHA-L analysis of ascending projections of the dorsal glutamatergic and GABAergic neurotransmission: state-dependent re- raphe nucleus in the rat. J Comp Neurol 313:643–668. sponse patterns observed in vitro. Neuroscience 141:2059–2066.