Distribution of mRNAs encoding CRF receptors in brain and pituitary of rat and mouse

Distribution of mRNAs Encoding CRFReceptors in Brain and Pituitary of Rat

and Mouse

KASIA VAN PETT,1 VICTOR VIAU,1 JACKSON C. BITTENCOURT,1

RAYMOND K.W. CHAN,1 HUI-YUN LI,1 CARLOS ARIAS,1 GAIL S. PRINS,3

MARILYN PERRIN,2 WYLIE VALE,2AND PAUL E. SAWCHENKO1*

1Laboratory of Neuronal Structure and Function, The Salk Institute for Biological Studiesand Foundation for Medical Research, La Jolla, California 92037

2Peptide Biology Laboratory, The Salk Institute for Biological Studies and Foundation forMedical Research, La Jolla, California 92037

3Department of Urology, University of Illinois College of Medicine, Chicago, Illinois 60612

ABSTRACTTwo G protein-coupled receptors have been identified that bind corticotropin-releasing

factor (CRF) and urocortin (UCN) with high affinity. Hybridization histochemical methodswere used to shed light on controversies concerning their localization in rat brain, and toprovide normative distributional data in mouse, the standard model for genetic manipulationin mammals. The distribution of CRF-R1 mRNA in mouse was found to be fundamentallysimilar to that in rat, with expression predominating in the cerebral cortex, sensory relaynuclei, and in the cerebellum and its major afferents. Pronounced species differences indistribution were few, although more subtle variations in the relative strength of R1 expres-sion were seen in several forebrain regions. CRF-R2 mRNA displayed comparable expressionin rat and mouse brain, distinct from, and more restricted than that of CRF-R1. Majorneuronal sites of CRF-R2 expression included aspects of the olfactory bulb, lateral septalnucleus, bed nucleus of the stria terminalis, ventromedial hypothalamic nucleus, medial andposterior cortical nuclei of the amygdala, ventral hippocampus, mesencephalic raphe nuclei,and novel localizations in the nucleus of the solitary tract and area postrema. Several sitesof expression in the limbic forebrain were found to overlap partially with ones of androgenreceptor expression. In pituitary, rat and mouse displayed CRF-R1 mRNA signal continu-ously over the intermediate lobe and over a subset of cells in the anterior lobe, whereasCRF-R2 transcripts were expressed mainly in the posterior lobe. The distinctive expressionpattern of CRF-R2 mRNA identifies additional putative central sites of action for CRF and/orUCN. Constitutive expression of CRF-R2 mRNA in the nucleus of the solitary tract, andstress-inducible expression of CRF-R1 transcripts in the paraventricular nucleus may pro-vide a basis for understanding documented effects of CRF-related peptides at a loci shownpreviously to lack a capacity for CRF-R expression or CRF binding. Other such “mismatches”remain to be reconciled. J. Comp. Neurol. 428:191–212, 2000. © 2000 Wiley-Liss, Inc.

Indexing terms: corticotropin-releasing factor; stress; neuropeptide receptors; pituitary gland;

urocortin

Dr. Van Pett’s current address is: The NeuroSurgery Group, Eugene,OR

Dr. Viau’s current address is: Department of Anatomy, University ofBritish Columbia, Vancouver, British Columbia, Canada.

Dr. Bittencourt’s current address is: Department of Anatomy, Universityof Sao Paulo, Brazil.

Dr. Li’s current address is: Department of Anatomy, Chang-Gung Uni-versity, Tao-Yuan, Taiwan.

Grant sponsor: NIH; Grant number: DK-26741; Grant sponsor: Foun-dation for Research and the Foundation for Medical Research; Grantsponsor: Medical Research Council of Canada; Grant sponsor: FAPESP;Grant number: 99/00213-5; Grant sponsor: American Heart Association-California Affiliate; Grant number: 95-117.

*Correspondence to: Dr. Paul E. Sawchenko, The Salk Institute, 10010N. Torrey Pines Rd., La Jolla, CA 92037. E-mail: [email protected]

Received 12 May 2000; Revised 1 August 2000; Accepted 11 August 2000

THE JOURNAL OF COMPARATIVE NEUROLOGY 428:191–212 (2000)

© 2000 WILEY-LISS, INC.

Corticotropin-releasing factor (CRF) is a 41-residue hy-pophysiotropic peptide, best known for its capacity tostimulate the synthesis and secretion of adrenocortico-

tropin (ACTH), and thus initiate pituitary-adrenal re-sponses to stress (see Chadwick et al., 1993). The peptidealso exhibits a broad distribution in brain (e.g., Merch-

ABBREVIATIONS

1–6 cortical layersA anterior lobe, pituitary glandACA anterior cingulate areaAHA anterior hypothalamic areaAMB nucleus ambiguusAMYG amygdalaAOB accessory olfactory bulbAON anterior olfactory nucleusAP area postremaAPT anterior pretectal nucleusARH arcuate nucleus (hypothalamus)BLA basolateral nucleus (amygdala)BST bed nucleus of the stria terminalisBSTe bed nucleus of the stria terminalis, encapsulated partBSTp bed nucleus of the stria terminalis, posterior divisionCA1 field CA1, Ammon’s hornCA3 field CA3, Ammon’s hornCBL cerebellumcc corpus callosumCER cerebellumCG central grayCLA claustrumCOAp cortical nucleus amygdala, posterior partCOCH cochlear nucleiCP choroid plexuscp cerebral peduncleCPu caudoputamenCRF corticotropin-releasing factorCtx cerebral cortexCU cuneate nucleusDBB nucleus of the diagonal bandD. COL. dorsal column nucleiDEEP N. deep nuclei (cerebellum)DG dentate gyrusDLL dorsal nucleus of the lateral lemniscusDMX dorsal motor nucleus vagus nerveDN dentate nucleusDR dorsal nucleus rapheec external capsuleEC external cuneate nucleusENd dorsal endopiriform nucleusENTl entorhinal area, lateral partENTm entorhinal area, medial partep endopiriform nucleusFRP frontal polegl glomerular layer (olfactory bulb)GP globus pallidusGr gracile nucleusgr granule cell layerHF hippocampal formationI intermediate lobe, pituitary glandIAD interanterodorsal nucleus (thalamus)IC inferior colliculusI. COL. inferior colliculusIII oculomotor nucleusIP interposed nucleusISO isocortexLA lateral nucleus amygdalaLC locus coeruleusLDT laterodorsal tegmental nucleusLGN lateral geniculate nucleusLHA lateral hypothalamic areaLRN lateral reticular nucleusLSd dorsal lateral septal nucleusLSi intermediate lateral septal nucleusLSv ventral lateral septal nucleusLV lateral ventricleMaPO magnocellular preoptic nucleusMB mammillary bodyme median eminenceMeA medial nucleus amygdalaMeAp medial nucleus amygdala, posterior part

MeV mesencephalic nucleus of the trigeminal nerveMGm medial geniculate complex, medial partmi mitral cell layer, olfactory bulbMOB main olfactory bulbmol molecular layer (cerebellum)mp medial parvicellular part (paraventricular nucleus)MPN medial preoptic nucleusMR median raphe nucleusMS medial septal nucleusMV medial vestibular nucleusND nucleus of DarkeschewitzNLOT nucleus of the lateral olfactory tractNTS nucleus of the solitary tractNTSc nucleus of the solitary tract, central subnucleusNTSd nucleus of the solitary tract, dorsal subnucleusNTSm nucleus of the solitary tract, medial subnucleusoch optic chiasmOB olfactory bulbOT olfactory tubercleP posterior lobe, pituitary glandp Purkinje cell layer, cerebellumPAR parasubiculumPBG parabigeminal nucleusPD posterodorsal preoptic nucleusPH posterior hypothalamic nucleusPir piriform areapm posterior magnocellular part (paraventricular nucleus)PG pontine grayPMd dorsal premammillary nucleusPP posterior lobe (pituitary gland)PPN pedunculopontine nucleusPRE presubiculumPRETECT pretectal regionPSV principal sensory nucleus of the trigeminal nervePT parataenial nucleusPVH paraventricular nucleus (hypothalamus)PVp posterior periventricular nucleus (hypothalamus)PVT paraventricular nucleus (thalamus)py pyramidal tractRM nucleus raphe magnusRT reticular nucleus (thalamus)SC superior colliculusscp superior cerebellar peduncleSept. septal regionS. COL. superior colliculusSF septofimbrial nucleusSN substantia nigraSNc substantia nigra, compact partSNV spinal trigeminal nucleusSNVc spinal trigeminal nucleus, caudal partSNVo spinal trigeminal nucleus, oral partSO supraoptic nucleusSOC superior olivary complexSUB subiculumSUBv ventral subiculumSUM supramammillary nucleusTRN tegmental reticular nucleusts solitary tractV3 third ventricleV4 fourth ventricleVCN ventral cochlear nucleusVENT THAL ventral thalamusVEST vestibular nucleiVII facial nucleusVLL ventral nucleus of the lateral lemniscusVMH ventromedial nucleus (hypothalamus)VTA ventral tegmental areaVTN ventral tegmental nucleuswm white matterXII hypoglossal nucleusZI zona incerta

192 K. VAN PETT ET AL.

enthaler et al., 1982; Swanson et al., 1983; Sakanaka etal., 1987; reviews: Sawchenko and Swanson, 1989;Sawchenko et al., 1993), and when administered centrallyevokes indices of autonomic and behavioral activation(e.g., Sutton et al., 1982; Brown and Fisher, 1985; reviews:Fisher, 1993; Koob et al., 1993). This has been taken assupporting an involvement of CRF in central stress-related functions that may complement its neuroendo-crine effects. A second member of the CRF peptide family,urocortin (UCN), has recently been identified (Vaughan etal., 1995), and has been found to display a central distri-bution distinct from, and more limited than, that of CRF(Kozicz et al., 1998; Bittencourt et al., 1999). Neverthe-less, evidence has been provided to suggest that centralactions of UCN may account for some stress-related ef-fects attributed originally to CRF (Spina et al., 1996).

Two G protein-coupled receptors have been identifiedthat bind CRF and/or UCN and activate adenylate cy-clase. The type 1 receptor (CRF-R1; Chang et al., 1993;Chen et al., 1993; Vita et al., 1993) binds both ligands withhigh affinity, with UCN being somewhat more potent thanCRF in activating through it, in vivo and in vitro(Vaughan et al., 1995; Asaba et al., 1998). This receptor isexpressed in pituitary in a manner compatible with theendocrine actions of CRF. In brain, however, the CRF-R1distribution is more suggestive of primary involvement incortical, cerebellar and sensory information processingthan in the stress-related visceromotor and behavioralresponses for which CRF is better known (Potter et al.,1994). A structurally related type 2 receptor (CRF-R2) hasbeen cloned by several groups from rodent tissue cDNAlibraries (Lovenberg et al., 1995a; Perrin et al., 1995); onevariant (CRF-R2b) is expressed in the periphery and thechoroid plexus, whereas a second (CRF-R2a) is expressedin brain (Lovenberg et al., 1995b), where its distribution isdistinct from, and more limited than, that of the R1 sub-type (Chalmers et al., 1995). CRF-R2 binds UCN in ahighly preferential manner; its affinity for CRF is low, asis the potency with which CRF signals through it (Loven-berg et al., 1995a; Vaughan et al., 1995).

Although general surveys (Potter et al., 1994; Chalmerset al., 1995), as well as more focused studies of the CRF-R1and R2 distributions in the rat model have been pub-lished, these have generally failed to identify clear sub-strates for CRF and/or UCN signaling at documented sitesof peptide action in eliciting stress-like effects (see, e.g.,Potter et al., 1994; Bittencourt and Sawchenko, 2000).Some localizations relevant to such effects remain contro-versial. For example, the paraventricular nucleus of thehypothalamus (PVH) has been identified as a sensitivesite at which local microinjections of CRF elicit appetite-suppressing and sympathomimetic effects (Brown, 1986;Krahn et al., 1988). The low basal level of CRF-R1 expres-sion at this locus has been found to be substantially in-creased in responses to diverse stressors (Luo et al., 1994;Rivest et al., 1995), although the relevance of inducibleexpression to the ability to respond to an acute challengeremains open to question. Moreover, the literature is con-flicting as to whether and under what conditions CRF-R2may be expressed in the PVH (Chalmers et al., 1995;Mansi et al., 1996; Lee and Rivier, 1997; Makino et al.,1998).

The purpose of this study was to expand upon the initialsurveys of CRF-R1 and -R2 expression in rat brain, par-ticularly as regards regions of ligand-receptor mismatch

(Herkenham, 1987) in documented sites of peptide action,and to extend the analysis to the mouse, the standardmammalian model for assessing the effects of genetic ma-nipulations. Mouse strains bearing targeted null muta-tions of CRF (Muglia et al., 1995), CRF-R1 (Smith et al.,1998), CRF-R2 (Bale et al., 2000; Coste et al., 2000; Kishi-moto et al., 2000) and the CRF-binding protein (Karolyi etal., 1999) have been generated, and display phenotypesthat have already begun to clarify the roles of these CRF-related signaling molecules in components of stress-related circuitry. Understanding of the normative distri-bution would seem an essential prerequisite forevaluating the effects of such manipulations. To this end,we have used hybridization histochemical methods tocompare and contrast the mRNA expression patterns ofthe two receptors in brain and pituitary of rat and mouse.

MATERIALS AND METHODS

Animals

Adult male and female Sprague-Dawley rats weighing250–350 g, and mice of the C57 BL/6 and NIH Swissstrains weighing 25–40 g, were used in this study, andhoused 2–3 per cage in a vivarium maintained on a 12:12hour light:dark cycle (lights on at 0600 hours). All animalshad free access to food and water at all times, and wereallowed a minimum of 7 days’ adaptation to housing con-ditions prior to any manipulation or killing for histology.All procedures were approved by the Institutional AnimalCare and Use Committee of the Salk Institute.

Tissue processing and histology

Animals were deeply anesthetized with chloral hydrate(35 mg/kg, i.p.) and perfused via the ascending aorta withsaline followed by 4% paraformaldehyde (pH 7.4) in 0.1 Mborate buffer (pH 9.5 at 10°C); 500–750 ml (for rats) or100–150 ml (for mice) of fixative was delivered over ;20minutes by using a peristaltic pump (Cole-Parmer, Ver-non Hills, IL). In most experiments, brains and pituitarieswere postfixed overnight at 4°C in fixative with 10% su-crose added. In experiments where some of the materialwas destined for use in both immuno- and hybridizationhistochemistry, postfixation times were reduced to 3hours, and tissue blocks were then transferred to 10%sucrose in 0.1 M phosphate buffer overnight at 4°C. Seriesof sections to be used for in situ hybridization were furtherpostfixed in phosphate-buffered 4% paraformaldehydeovernight at 4°C before being transferred to cryopro-tectant for storage.

Tissue blocks were frozen in dry ice and multiple regu-larly spaced series of 30-mm-thick sections in the trans-verse plane were cut and saved. For rats, sampling inter-vals were 150 or 300 mm, whereas in mice this wasreduced to 90–120 mm. One-in-10 series sections werecollected in cold cryoprotectant (0.05 M sodium phosphatebuffer, 30% ethylene glycol, 20% glycerol), and stored at-20°C until histochemical processing.

Probes

Multiple probes were screened for each receptor in eachspecies. Those that provided optimal results, and on whichthe bulk of our observations were based, are listed below.For CRF-R1 mRNA localization in rat, radiolabeled anti-sense and sense (control) cRNA copies were synthesized

193CRF RECEPTOR DISTRIBUTIONS IN RAT AND MOUSE

from a full-length (1.3 kb) rat CRF-R1 cDNA (Potter et al.,1994) subcloned into a pBluescript SK transcription vector(Stratagene, La Jolla, CA). CRF-R1 probes in mouse weregenerated from a 1.2-kb cDNA encompassing the entirecoding region of the receptor obtained by reversetranscriptase-polymerase chain reaction (RT-PCR) ampli-fication from AtT-20 cells by using primers based on thehuman brain R1 sequence (Vita et al., 1993), which dis-plays 97% amino acid sequence identity with mouse.

Similar, and equally good results were obtained in ratand mouse by using a probe encompassing 0.9 kb of thecoding sequence and 0.1 kb of 5’ untranslated region ofmouse CRF-R2b which was adjusted to an average frag-ment length of ;200 bases by limited alkaline hydrolysis(Cox et al., 1984) prior to application to tissue sections.Use of this probe enabled both CRF-R2 RNA processingvariants (Lovenberg et al., 1995a) to be detected. Probesspecific for rat CRF-R2a (275 nt) and b (461 nt), gener-ously provided by Neurocrine Biosciences (San Diego, CA)(see Chalmers et al., 1995; Lovenberg et al., 1995a,b),provided similar results, although the strength of hybrid-ization signals obtained in brain by using the rat CRF-R2aprobe were weaker than those seen by using the digestedmouse probe.

In situ hybridization

Hybridization histochemical demonstration of CRF-R1and CRF-R2 mRNAs was carried out by using 33P- or35S-labeled antisense cRNA probes. Techniques for probesynthesis, hybridization, and autoradiographic localiza-tion of mRNA signal were adapted from Simmons et al.(1989). In brief, tissue processed as above was mountedonto poly-L-lysine-coated slides, and then digested with 10mg/ml of proteinase K for 30 minutes, at 37°C. ForCRF-R1 mRNA localization, radiolabeled antisense andsense (control) cRNA copies were synthesized from a full-length rat (1.3 kb) CRF-R1 cDNA (Potter et al., 1994)subcloned into a pBluescript SK transcription vector(Stratagene, La Jolla, CA).

The probes were used at concentrations of about 107

cpm/ml, and applied to sections overnight at 56–58°C in asolution containing 50% formamide, 0.3 M NaCl, 10 mMTris (pH 8.0), 1 mM EDTA, 0.05% tRNA, 10 mM dithio-threitol, 13 Denhardt’s solution, and 10% dextran sulfate,after which they were treated with 20 mg/ml of ribonucle-ase A for 30 minutes at 37°C, and washed in 15 mMNaCl/1.5 mM sodium citrate at 55–60°C. Sections werethen dehydrated and exposed to X-ray films for 1–2 days.Sections were defatted in xylene, rinsed in absolute etha-nol, air-dried, coated with Kodak NTB-2 liquid autoradio-graphic emulsion, and exposed at 4°C, dark and desic-cated, typically for 3–4 weeks. They were then developedwith Kodak D-19 for 3.5 minutes at 14°C, rinsed briefly indistilled water, fixed with film strength Kodak rapid fixerfor 2 minutes at 14°C, rinsed again, and counterstainedwith thionin for reference purposes.

Stress paradigms

To probe for inducible sites of CRF-R expression, mate-rial from rats perfused at varying intervals (1, 2, 3, 4, 6, or24 hours) after acute exposure to one of two stress para-digms, was used. These included hypotensive hemor-rhage, achieved by withdrawing an estimated 15% bloodvolume from jugular catheters implanted under methoxy-flurane anesthesia 2 days before the challenge, by using

methods detailed previously (Chan and Sawchenko, 1994).Another group of animals received a single session ofinescapable electrical footshock, in which 60 shocks (1 mAfor 1 msec) were delivered randomly over the course of a30-minute session in chambers to which the rats had beenadapted in the course of 7 daily 30-minute sessions duringwhich no current was delivered (Li and Sawchenko, 1998).

Immunohistochemistry

To provide an independent assessment of the effective-ness of stress procedures, immunolocalization of Fos-immunoreactivity (Fos-ir) was carried out. Sections werestained by using conventional avidin-biotin immunoper-oxidase methods (Sawchenko et al., 1990) to localize aprimary antiserum raised against a synthetic N-terminalfragment of human Fos (Santa Cruz Biotechnology, SantaCruz, CA), and used at a 1:10,000 dilution. Adjoiningseries of sections were stained with thionin for referencepurposes. Tests for specificity of immunolabeling involvedsubstitution of nonimmune serum for the primary anti-serum, or using primary antiserum that had been incu-bated overnight at 4°C with 50 mM of the synthetic im-munogen. Neither procedure gave rise to any suggestion ofspecific labeling in material from control or experimentalanimals.

Combined immuno- and hybridizationhistochemistry

Because major aspects of the CRF-R2 expression pat-tern appeared similar to those of gonadal steroid hormonereceptors, particularly the androgen receptor (see Simerlyet al., 1990), dual localization experiments were carriedout to compare the distributions directly. For this purpose,the androgen receptor immunoreactivity was localized byusing a polyclonal antiserum (PG-21) raised in rabbitagainst amino acids 1–21 of the rat androgen receptor(Prins et al., 1991). For dual localization experiments,immunostaining was carried out first, and the two constit-uent methods were modified as described elsewhere toallow efficient dual localization (Chan et al., 1993). Inbrief: (1) blocking serum was omitted from the immuno-staining procedure; these were replaced with 2% bovineserum albumin and 2% heparin sulfate, and (2) nickelenhancement steps were eliminated from the immuno-staining protocol. Localization was achieved by using theavidin-biotin-immunoperoxidase protocol by using Vec-tastain Elitet reagents. Sections were mounted onto glassslides and were then processed for in situ hybridizationhistochemistry as described above.

Imaging

Low-magnification darkfield macrophotographic imag-ing (Figs. 1, 7) was carried out by using a Nikon Multiphotsystem, Kodak Ektapan film, and were rendered photo-graphically. The edges of the prints and artifact over ven-tricles were retouched. All other images were captured oneither Ilford XP-2 or Kodak Ektachrome 160 film. Figures2, 4, and 5 were printed photographically, whereas imagescomprising Figures 3, 8 and 9 were digitized by using aKodak RS-3570 film scanner, imported into Adobe Photo-shop (v. 5.0), adjusted to balance and optimize brightness,contrast, and sharpness. Off-tissue backgrounds weredarkened for clarity. Individual files were exported toCanvas (v. 3.55) for assembly into plates, which were


rendered at a resolution of 300 dpi on a Kodak PS 8600dye sublimation printer.

RESULTS

CRF-R1 distribution in rat and mouse brain

We have described essential features of the distributionof CRF-R1 in rat brain and pituitary in a somewhat ab-breviated format (Potter et al., 1994), and the presentobservations are fully in accord with this account. Here wesimply provide a more comprehensive listing of the rela-tive strengths of R1 mRNA expression in rat brain regions(Table 1), and use this as a basis for comparison of thedistribution of transcripts encoding this receptor inmouse. These were found to be fundamentally similar,with some apparent differences in regional emphasis, andfewer ostensibly qualitative ones. Observations weremade on material from 21 rats (15 male, 6 female) and 14mice (10 male, 4 female). No clear differences in distribu-tion as a function of gender were apparent in either spe-cies.

Cortex. As in rat, isocortex comprised the dominantseat of CRF-R1 mRNA expression in mouse forebrain (Fig.1). Labeling was most dense over layer 4, with secondaryaccumulations seen over layers 2/3. The moderate densityof labeled cells seen in layer 6 of rat isocortex was notevident in mouse, where labeling density was comparableto that seen in layer 5 (Fig. 2), which bore the lowestdensity of R1-expressing cells in rat cortex. In neitherspecies were systematic differences seen in either the lam-inar distribution or relative prominence of positively hy-bridized cells across isocortical fields.

CRF-R1 mRNA expression in mouse hippocampal for-mation was similar to that in rat, in displaying a moderatesignal continuously over the pyramidal cell layers of Am-mon’s Horn and the subicular complex. In addition, amoderate number of positively hybridized cells were local-ized principally to the hilar region of the dentate gyrus.

In both species, type 1 receptor mRNA expression waspervasive throughout the olfactory system. All cellularsubregions of the main and accessory olfactory bulbs wereoverlain by hybridization signals of at least moderate in-tensity. Much the same was true of downstream levels ofthe main olfactory pathway, including the anterior olfac-tory nucleus, olfactory tubercle, piriform cortex, andolfactory-related parts of the amygdala.

Subcortical telencephalic areas. In contrast to therat caudoputamen, which displayed a homogeneous low-level mRNA signal (see also Potter et al., 1994), R1 ex-pression in mouse showed a decidedly patchy distributionwith small clusters of cells showing moderately stronglabeling (Fig. 1). As in rat, robust signals were seen infunctionally associated cell groups, including the magno-cellular preoptic and subthalamic nuclei, the substantianigra, and the ventral tegmental area.

In the septal region of the mouse, CRF-R1 mRNA wasexpressed preferentially in the medial septal nucleusand nucleus of the diagonal band, although on a relativebasis both the labeling intensity and the density ofpositively hybridized cells were clearly lower than thoseseen reliably in rat (Figs. 1, 2). In both species, mostrecognized nonolfactory portions of the amygdala dis-played low to moderate intensity hybridization signals,with those over the basolateral and medial nuclei stand-

ing out as being somewhat more prominent. The notableexception to this generalization was the central nucleus;whereas both rat and mouse displayed a diffuse low-level mRNA signal over the medial part of the centralnucleus, the lateral part was conspicuously devoid ofabove-background labeling.

Thalamus. The patterns of CRF-R1 expression in thethalamus of rat and mouse were quite distinct. Althoughboth displayed low to moderate level signals over severalmidline nuclei and aspects of the lateral geniculate com-plex, we found no counterpart in mouse for the low tomoderate R1 signals seen over the lateral and ventralnuclear groups in rat. Instead, the dominant site of recep-tor expression in mouse thalamus was the reticular nu-cleus (Fig. 1), which was never observed to display above-background hybridization signals in rat.

Hypothalamus. CRF-R1 signal over most recognizedcell groups of the mouse hypothalamus also paralleled thesituation in rat in being generally weak and diffuse.Among the cell groups that stood out as displaying moreprominent R1 expression were the supramammillary nu-cleus, the tuberal part of the lateral hypothalamic area,and the arcuate nucleus, which, relative to adjoining hy-pothalamic cell groups, was more strongly and discretelylabeled in mouse than in rat (see Figs. 1, 2). The paraven-tricular nucleus of the hypothalamus contained a sparseto equivocal signal in both species, which in rat localizedprincipally to autonomic-related aspects of the parvicellu-lar division. In neither rat nor mouse were above-background R2 mRNA signals detected over the supraop-tic nucleus.

Brainstem regions. As in rat, many of the moreprominent sites of CRF-R1 mRNA expression in mousebrainstem conformed to sensory relay structures. Suchincluded cell groups involved in the processing of somaticsensory information, such as the dorsal column, peduncu-lopontine, laterodorsal tegmental nuclei, and all majortrigeminal sensory structures (Fig. 1). Each of the vestib-ular nuclei contained a moderate density of positivelyhybridized cells, as did multiple levels of the primaryauditory pathway, including the ventral cochlear, lateralsuperior olivary, lateral lemniscal nuclei, and inferior col-liculus. Lower level labeling was seen over aspects of themedial division of the medial geniculate complex. In addi-tion to the visual cortex and thalamus, CRF-R1 was ex-pressed at moderate levels in the superior colliculus, in-cluding the superficial gray, and in several cell groups inthe visual pretectum. Among visceral sensory structures,the most prominent site of R1 expression was the para-brachial nucleus, including both its lateral (interoceptive)and medial (gustatory) divisions. Among subregions of theparabrachial nucleus (see Fulwiler and Saper, 1984), theexternal lateral subnucleus displayed the highest densityof positively hybridized cells, with the central lateral sub-nucleus containing fewer, although more intensely la-beled, neurons. Among recognized subregions of the nu-cleus of the solitary tract, only the central subnucleus(Ross et al., 1985), a cell group associated with esophagealmotor function (Cunningham and Sawchenko, 1989), wasassociated in both species with a discrete CRF-R1 mRNAsignal of any prominence (see Fig. 8).

Cerebellum and associated cell groups. Observa-tions made in rat and mouse cerebellum were consonantin revealing moderate signal intensities over the Purkinjeand granule layers of cerebellar cortex. Each of the deep


TABLE 1. Major Sites of Corticotropin-Releasing Factor (CRF) Receptor Subtype mRNA Expression in Rat and Mouse Brain1

CRF-R1 CRF-R2

Rat Mouse Rat Mouse

I. ForebrainA. Isocortex

II 11 11 2 2III 11 11 2 2IV 111 111 2 2V 1 1 2 1VI 11 1 1 1Claustrum 11 1 1 11

B. Olfactory regions1. Main bulb

Olfactory nerve layer 2 2 2 2Periglomerular layer 11 11 2 2Int. plexiform layer 1 2 2 2Mitral layer 111 111 2 2Ext. plexiform layer 11 1 2 2Granule cell layer 111 111 111 11

2. Anterior olfactory n. 11 11 1 13. Olfactory tubercle 11 11 1 114. Piriform cortex 111 11 1 1

Endopiriform nucleus 111 2 1 115. Taenia tecta 111 111 2 2

C. Hippocampal formation1. Entorhinal area (lateral and medial) 11 11 11 112. Subiculum (dorsal) 11 11 1 11

Subiculum (ventral) 2 2 1 113. CA1 11 11 1 14. CA3 11 11 1 15. Dentate gyrus

Granular layer 1 2 1 1Polymorph layer 11 11 2 2

6. Induseum griseum/fasciola cinerea 2 2 2 2D. Amygdala

1. Medial nucleusAnterior part 1 11 11 11Posterodorsal part 11 1 11 111

2. Amygdalohippocampal area 11 1 2 23. Cortical nucleus

Anterior part 11 11 11 11Posterior part 11 11 111 11

4. N. lat. olfactory tract 1 11 1 115. Anterior amygdaloid area 11 11 2 26. Central nucleus (lat./med.) 2/1 2/1 2/2 2/27. Lateral nucleus 1 2 2 18. Basolateral nucleus 111 1 1 19. Basomedial nucleus 11 1 11 1110. Intercalated nuclei 1 2 2 2

E. Septum1. Lateral nucleus

Dorsal part 2 2 11 1Intermediate part 2 1 1111 1111Ventral part 1 2 2 2

2. Medial n./N. diagonal band 111 11 2 113. Bed n. Stria terminalis 11 11

Rostromedial region 11 1 2 2Rostrolateral region 11 1 2 2Posterodorsal region 111 11 11 111Posteroventral region 11 1 1 1

4. Bed n. anterior commissure 1 2 2 25. Septofimbrial nucleus 1 2 11 116. Triangular nucleus 1 2 11 117. Subfornical organ 1 2 2 2

F. Basal ganglia1. Caudoputamen 11 11 2 2

Posteroventral part 1 2 2 2Nucleus accumbens 11 1 2 2Fundus of striatum 1 2 1 1

2. Globus pallidus 11 111 2 2Entopeduncular n. 1 2 2 2Substantia innominata 11 1 2 2Magnocellular preoptic nucleus 111 11 1 1

3. Subthalamic nucleus 111 11 2 24. Substantia nigra

Compact part 111 11 2 2Reticular, lateral parts 11 1 2 2Ventral tegmental area 11 1 2 2

G. Thalamus1. Medial habenula 2 2 2 22. Lateral habenula 2 2 2 23. Anterior group

Anteroventral n. 2 2 2 2Anteromedial n. 2 2 2 2Anterodorsal n. 2 2 2 2

4. Mediodorsal nucleus 1 2 2 25. Lateral group

Lateral dorsal n. 11 2 2 2Lateral posterior n. 11 2 2 2


CRF-R1 CRF-R2

Rat Mouse Rat Mouse

6. Midline groupParaventricular n. 11 1 2 2Parataenial n. 2 11 2 2N. reuniens 1 2Rhomboid n. 2 2 2 2N. gelatinosa 2 2 2 2

7. Ventral group 2 2 2Ventral anterior/v. lat. 11 1 2 2Ventral medial 11 2 2 2Ventral posterior 11 1 2 2Gustatory nucleus 1 2 2 2

8. Posterior complex 11 2 2 29. Medial geniculate n.

Dorsal part 1 2 2 2Ventral part 1 2 2 2Medial part 11 11 2 2

10. Lateral geniculate n.Dorsal part 1 2 2 2Intergeniculate leaflet 1 11 2 2Ventral part 111 11 2 2

11. Intralaminar nucleiCentral medial n. 2 2 2 2Paracentral n. 1 2 2 2Central lateral n. 1 2 2 2Parafascicular n. 1 2 2 2

12. Reticular nucleus 2 111 2 213. Zona incerta

Rostral 11 11 2 2Caudal 11 1 2 2

14. N. fields of Forel 1 2 2 2H. Hypothalamus

1. Periventricular zoneMedian preoptic n. 1 2 2 2Anteroventral periventricular n. 11 2 2 2Preoptic periventricular nucleus 11 1 2Suprachiasmatic n. 11 1 2 2Supraoptic nucleus 11 2 1 1Paraventricular n. 1 1

Autonomic part 1 2Parvicellular part 2 2Magnocellular part 2 1

Anterior periventricular n. 1 2 2 2Arcuate nucleus 11 11 1 1Posterior periventricular n. 1 2 1 11

2. Medial zoneMedial preoptic area 11 2 1 1Medial preoptic n.

Lateral part 1 2 1 1Medial part 11 1 11 11Central part 1 2 2 2

Anterior hypo. n.Anterior part 1 2 1 1Central, posterior parts 1 1 1 1

Retrochiasmatic area 1 2 2 2Ventromedial n. 2 2 111 11Dorsomedial n. 111 1 1 1Tuberomammillary n. 2 2 2 2Premammillary n.

Dorsal part 1 2 1 11Ventral part 1 1 2 2

Supramammillary n.Lateral part 11 11 2 2Medial part 11 11 2 1

Lateral mammillary n. 11 1 2 2Medial mammillary n. 2 2 2 2

3. Lateral zoneLateral preoptic area 1 1 1 1Lateral area. 1 1 1 11Posterior area 111 11 2 1

I. Pituitary Gland1. Anterior lobe 11 11 2 22. Intermediate lobe 11 11 2 23. Posterior lobe 2 2 11 11

II. BrainstemA. Sensory

1. VisualSuperior colliculusSuperficial gray 11 11 2 2Intermediate gray 1 1 2 1

Parabigeminal n. 1 11 2 2Pretectal region

Olivary n. 11 2 2 2N. optic tract 1 2 2 2Anterior n. 11 2 2 2Posterior n. 1 2 2


TABLE 1. (continued)

CRF-R1 CRF-R2

Rat Mouse Rat Mouse

Medial pretectal a. 11 2 1 2N. posterior commissure 1 2 2 2

Medial terminal n. 1 2 2 22. Somatosensory

Mesencephalic n. V 1 2 1 11Principal sensory n. V 111 1111 2 2Spinal n. V 111 11 2 1

Oral part 111 111 2 2Interpolar part 111 111 2 2Caudal part 1111 1111 1 1

Dorsal column n. 1111 1111 2 2External cuneate n. 1111 1111 2 2

3. AuditoryCochlear nuclei

Dorsal 111 11 2 2Ventral 11 1 2 2

N. trapezoid body 1 1 2 2Superior olive-lateral 11 1 2 2N. lateral lemniscus

Ventral 11 11 2 2Dorsal 1 1 2 2

Inferior colliculusExternal 1 1 11 1Dorsal 1 1 1 2Central 11 11 1 1

N. brachium inf. coll. 2 2 2 2N. saguluum 2 2 2 2

4. VestibularMedial n. 11 11 2 2Lateral n. 11 11 2 2Superior n. 11 11 2 2Spinal n. 11 1 2 2

5. GustatoryN. solitary tract, ant. 1 2 2 2

6. VisceralN. solitary tract

Medial division 1 1 11 11Commissural division 2 2 2 2Lateral division 1 1 2 2

Area postrema 2 2 11 11Parabrachial n.

Lateral 111 111 2 2Medial 11 11 2 2Kolliker-Fuse n. 1 1 2 2

B. Motor1. Eye

Oculomotor (III) 11 2 2 2Trochlear (IV) 2 2 2 2Abducens (VI) 1 1 2 2

2. JawMotor n. V 1 2 2 2

3. FaceFacial n. (VII) 11 1/11 2 2

4. Pharynx/larynxN. ambiguus 11 11 2 2

5. Tongue 2 2Hypoglossal n. (XII) 2 2 2 2

6. VisceraDorsal motor n. X 1 11 2 2

C. Reticular core (including central gray)1. Periaqueductal gray 1 1 1 1

Interstitial nucleus of Cajal 1 1 2 2N. Darkschewitsch 111 11 2 2Dorsal tegmental n. 2 2 2 2Ventral tegmental n. 11 11 2 2N. incertus 1111 1 2 2Laterodorsal teg. n. 1111 11 2 2Barrington’s n. 2 2 2 2Locus coeruleus 2 2 2 2

2. RapheInterfascicular n. 2 2 2 2Rostral linear n. 1 1 2 2Dorsal raphe 1 2 111 111Median raphe 11 2 11 11N. raphe pontis 2 1 2 2N. raphe magnus 11 1 2 2N. raphe obscurus 11 1 2 2N. raphe pallidus 1 2 2 2

3. Interpeduncular n. 11 1 111 114. Reticular formation

Central teg. fieldSubcuneiform part 1 1 2 2Retrorubral part 2 1 2 2

Peripeduncular n. 1 2 2 2Pedunculopontine n. 1111 11 2 2



cerebellar nuclei were strongly and uniformly positive forR1 transcripts, as were a number of major pre- and post-cerebellar structures, including the red, lateral reticularand basilar pontine nuclei. Lesser signal intensities wereseen in each species over the perihypoglossal nuclei,whereas the inferior olive displayed little or no evidence ofCRF-R1 expression.

Pituitary gland. As described previously in rat (Pot-ter et al. 1994), mouse pituitary consistently displayedrobust hybridization signals for CRF-R1 mRNA through-out the intermediate lobe, and over a subset of anteriorpituitary cells (Fig. 3). Above-background labeling was notobserved over the posterior lobe of the gland.

Inducible sites of CRF-R1 expression. Severalgroups have described inducible expression of CRF-R1 inthe paraventricular nucleus of the hypothalamus in ratfollowing exposure to a variety of different stress para-digms (Luo et al., 1994; Rivest et al., 1995). We haveexposed separate groups of rats to one of two additionalmodels, hypotensive hemorrhage or electrical footshock, totest the generality of this phenomenon, to determinewhether such changes are limited to the PVH, and tofollow the time course of observed changes in receptorexpression.

Both manipulations gave rise to induced expression ofCRF-R1 transcripts in the PVH. This generally conformedto the pattern of cellular activation (Fos-ir induction) seenin response to these stress paradigms in being localizedprimarily to the medial parvicellular part of the nucleus,with secondary involvement of the magnocellular divisionand the supraoptic nucleus in tissue from hemorrhagedrats (Fig. 4). Induction in both paradigms followed a rel-atively slow time course in being first detectable at 2hours, and maximal at 4–6 hours, after stress.

Examination of extrahypothalamic cell groups revealedno additional sites of stress-induced CRF-R1 expression.Thus, despite robust Fos induction seen in response to oneor both challenges in other cell groups implicated as sitesof CRF action in eliciting stress-related responses, no ev-

idence for CRF-R1 expression was adduced at any post-stress time point (Fig. 5).

CRF-R2 distribution in rat and mouse brain

Observations on the distribution of CRF-R2 mRNA inrat brain (Fig. 6) were generally in good agreement withthose reported by Chalmers et al. (1995), and includedmajor sites of expression in the granule cell layers of themain and accessory olfactory bulbs, aspects of the lateralseptal nucleus, and the bed nucleus of the stria terminalis,the medial and posterior cortical nuclei of the amygdala,the ventromedial nucleus of the hypothalamus, the mes-encephalic raphe, and interpeduncular nuclei, and non-neuronal elements of the choroid plexus. Our material,however, supported a number a differences in emphasis,as well as several novel localizations, including a poten-tially important one for reconciling ligand-receptor mis-matches in documented sites of peptide action, in thenucleus of the solitary tract. These departures will beconsidered below in the context of describing CRF-R2mRNA expression patterns in mouse brain.

Cortex. Although not rivaling the widespread corticaldistribution of CRF-R1, type 2 receptor mRNA was foundreliably to be expressed in cortex of both species. In iso-cortex, this was manifest mainly as scattered positivelyhybridized neurons in deeper layers (5 and 6), which weremost numerous in temporal areas (Figs. 2, 7). In thehippocampal formation, a sparse to equivocal signal wasseen over the principal cell layers of both the dentategyrus and Ammon’s horn, which tended to be more robustin mouse than in rat brain. Interestingly, however, weobserved substantially more robust labeling over aspectsof the retrohippocampal cortical fields, most prominentlyincluding layer 2 of the pre- and parasubiculum and en-torhinal cortex. The most salient site of CRF-R2 expres-sion in the olfactory system was a moderately strong sig-nal over the internal part of the granule cell layer of boththe main and accessory olfactory bulbs.

CRF-R1 CRF-R2

Rat Mouse Rat Mouse

Cuneiform n. 2 1 2 2Pontine reticular 11 1 2 2Linear n. medulla 11 11 2 2Parvicellular ret. field 1 11 2 2Gigantocellular ret. field 11 1 2 2Lateral paragigantocellular

Intermediate ret. field 2 1 2 2Paramedian reticular n. 1 1 2 2

D. Pre- and Postcerebellar1. Pontine gray 1111 1111 2 22. Tegmental reticular n. 11 1 2 23. Inferior olive 1 2 2 24. Lateral reticular n. 1111 111 2 25. Red nucleus 111 11 2 26. N. Roller 1 2 2 27. N. Prepositus 111 111 2 2

III. CerebellumA. Deep nuclei 111 111 2 2B. Cortex

Molecular layer 11 2 2 2Purkinje layer 11 1 2 2Granule layer 111 11 2 2

1The relative strength of expression of each transcript in each cell group was rated by two independent observers. Ratings reflect primarily the density of positively labeled cells,with (2) representing a complete lack of above-control levels of labeling, (1) isolated positively labeled cells, and (1111) labeling in a substantial majority of all cells in a givencell group or field. Ratings of mRNA expression were adjusted secondarily on the basis of the strength of hybridization signal, but never by more than a single rating point.



Subcortical telencephalon. As in rat, major sites ofCRF-R2 expression in mouse brain were evident in theseptal region and amygdala (Fig. 7). Dominant amongthese was a large number of positively hybridized cells inthe ventrolateral part of the intermediate lateral septalnucleus. A substantially lower level of expression overaspects of the dorsal lateral septal nucleus was seen in rat,

but not in mouse, material. In stark contrast to the rat,however, where R1 and R2 transcripts segregated quitediscretely to the medial and lateral septal nuclei, respec-tively (see Vaughan et al., 1995; Bittencourt andSawchenko, 2000), R2 transcripts were also detected inthe medial septal/diagonal band complex of the mouse,with a prominence that rivaled that of R1 expression (Fig.

Fig. 1. A–I: Hybridization histochemical localization of corticotropin-releasing factor type 1 receptor (CRF-R1) mRNA in the mouse brain.Darkfield photomicrographs arranged from rostral (A) to caudal(I) show the regional distribution of neurons positively hybridizedwith an antisense cRNA probe generated from a full-length mouseCRF-R1 cDNA. The general pattern of expression is similar to that

seen in rat, with a few significant exceptions. These include thethalamus, which is dominated in mouse by labeling over the reticularnucleus (RT), and the basal ganglia, where the caudoputamen (CP)shows patchy CRF-R1 expression. For abbreviations in this and sub-sequent figures, see list. Scale bar 5 2 mm.


2). In addition, CRF-R2-expressing cells were detected inboth species’ caudal (septofimbrial and triangular) septalnuclei.

The posteromedial aspect of the bed nucleus of the striaterminalis comprised another major site of CRF-R2mRNA expression, which tended to be somewhat more

salient in mouse than in rat brain. In both species, hybrid-ization signal was concentrated over a region designatedby Ju and Swanson (1989) as the encapsulated part of thebed nucleus, and was contiguous with a smaller groupingof labeled cells in the posterodorsal preoptic nucleus (seeSimerly et al., 1990).

Fig. 2. Comparison of corticotropin-releasing factor type 1 recep-tor (CRF-R1) and type 2 receptor (CRF-R2) expression patterns inmouse brain regions. Sections through comparable levels of the sep-tum (top), basomedial hypothalamus (middle), and auditory cortex(bottom), hybridized with probes for CRF-R1 (left) and CRF-R2 (right)mRNA. Both receptors are expressed at low to moderate levels in themedial septal nucleus (MS), whereas the R2 transcript is massively

expressed in the lateral septum (LS). Both receptor mRNAs are ex-pressed in the arcuate nucleus and lateral hypothalamic area, al-though with distinctive topographies and strengths; only CRF-R2mRNA is expressed in the ventromedial nucleus. In neocortex,CRF-R1 signal over layers 2/3 and 4 predominates, with only a fewscattered cells in the deeper layers displaying R2 mRNA. Scale bars 5500 mm in top two panels; 250 mm in bottom four.


In the mouse amygdala, CRF-R2 expression was focusedin the medial nucleus, being particularly robust in itsposterior aspect (Fig. 7), and in the basomedial nucleus.The posterior cortical nucleus, a dominant site of R2 ex-pression in rat, comprised only a secondary locus of recep-tor expression in mouse amygdala. Reliable labeling wasnot detected in either species over any aspect of the cen-tral nucleus.

Thalamus. No clear evidence was obtained forCRF-R2 expression in the thalamus of either rat or mouse.

Hypothalamus. In rat, the dorsomedial part of theventromedial nucleus stood as the clearly dominant hypo-thalamic site of CRF-R2 expression. Although a compara-ble localization was evident in mouse brain, the strengthof signal was comparable to that seen over the lateralaspect of the arcuate nucleus, the dorsal premammillarynucleus, and more widely scattered cells in the medialpreoptic and lateral hypothalamic areas (Fig. 2).

The questions of whether CRF-R2 is expressed in thePVH, and if so, in which functional compartments, hasremained somewhat controversial (cf Chalmers et al.,

1995; Mansi et al., 1996; Lee and Rivier, 1997; Makino etal., 1998). In our better preparations of both rat andmouse brain, we observed relatively weak, althoughclearly above-background, hybridization signals over as-pects of the PVH. In rat, which displays a more pro-nounced topographic organization of major output neuronclasses (Swanson and Sawchenko, 1983), these wereclearly localized to the magnocellular division of the nu-cleus (Fig. 8). Accordingly, comparably muted R2 mRNAsignal was seen in both species over the supraoptic nu-cleus, the other principal focus of the magnocellular neu-rosecretory system.

Cerebellum. In none of our experiments was evidenceobtained for CRF-R2 expression in the cerebellum of ei-ther species.

Brainstem. The mesencephalic raphe nuclei and as-pects of the interpeduncular nuclear complex comprisedthe dominant sites of CRF-R2 mRNA expression in thebrainstem of both rat and mouse. In addition, our obser-vations supported previous reports of less prominent (interms of both cell number and labeling intensity) localiza-tions in the superior and inferior colliculi, and in themarginal zone of the spinal trigeminal nucleus (Chalmerset al., 1995). In addition, we noted consistently in bothspecies hybridization signal associated with the mesence-phalic trigeminal nucleus and with a population of cells inthe medial part of the most rostral aspect of the midbrainperiaqueductal gray.

As noted above, consistent evidence was also obtainedfor a substantial expression of CRF-R2 in visceral sen-sory structures of the dorsomedial medulla. This in-cluded expression concentrated in the dorsal subnu-cleus of the nucleus of the solitary tract and adjoining(ventrolateral) aspects of the area postrema (Fig. 8; seealso Figs. 6, 7).

Pituitary gland. Both rat and mouse material dis-played a predominant localization of CRF-R2 transcriptsover the posterior lobe of the pituitary gland (Fig. 3). Theoverall signal intensity over the anterior lobe was greaterthan that seen in sections from the same animals hybrid-ized with sense-strand runoffs, but lacked crisp cellularresolution, thereby defeating any attempt at further char-acterization. Positive labeling was never detected over theintermediate lobe of the gland.

Non-neuronal elements. In line with previous work,probes specific for the long (b) form of the type 2 receptor(Chalmers et al., 1995; Lovenberg et al., 1995b) consis-tently revealed strong hybridization signals associatedwith the choroid plexus (Figs. 6, 7) and blood vessels at themargins of the brain. Such localizations were never seenby using CRF-R2a-specific probes. Our material failed toprovide support for a reported localization of CRF-R2 mes-sage to the ependymal lining of the ventricular system (cfChalmers et al., 1995).

Relation to sites of androgen receptor expression.

Aspects of the CRF-R2 distribution in the limbic forebrainwere strongly reminiscent of the distribution of gonadalsteroid hormone receptors, particularly that of the andro-gen receptor (AR; Simerly et al., 1990). To assess possibleoverlap, complete series of sections throughout the brainsof four adult male rats were prepared for concurrent duallocalization of AR-ir and CRF-R2 mRNA. Although thereduction in the strength of hybridization signal in mate-rial prepared for dual localization limited the scope of theanalysis to sites of more robust R2 expression, regions of

Fig. 3. Corticotropin-releasing factor (CRF) receptor mRNA distri-butions in mouse pituitary. Darkfield photomicrographs of sectionsthrough mouse pituitary showing hybridization signals for type 1receptor (CRF-R1) and type 2 receptor (-R2) mRNAs. The CRF-R1transcript is prominently expressed in the intermediate lobe (I), andover a subset of anterior lobe (A) cells, whereas R2 mRNA displays apredominant localization over the posterior lobe (P). In addition, anabove-background signal that generally lacks clear cellular resolutionis seen over the anterior lobe. Scale bar 5 500mm.


partial overlap were identified (Fig. 9). These included theencapsulated part of the bed nucleus and the posteriorpart of the medial nucleus of the amygdala, where diffuseR2 mRNA signal codistributed with dense accumulationsof AR-ir neurons. In the intermediate lateral septal nu-cleus and the ventromedial nucleus of the hypothalamus,however, the two distributions displayed clear tendenciestoward topographic segregation, with clear examples ofdoubly labeled cells limited largely to areas of partialoverlap. Among other potential sites of congruence, weregularly noted examples of doubly labeled cells in themedial preoptic area, but detected none in the nucleus ofthe solitary tract.

DISCUSSION

The results of the present study provide normative dataon the localization of CRF-R mRNAs in mouse brain andpituitary, and an expanded treatment of these receptors’distributions in rat tissues. A general summary of theobserved patterns of expression is provided in Figure 10.The major goals were to provide a context for evaluatingthe effects of genetic manipulations of CRF-related signal-ing molecules in the mouse model (Muglia et al., 1995;Smith et al., 1998; Karolyi et al., 1999; Bale et al., 2000;Coste et al., 2000; Kishimoto et al., 2000), and to attemptto define and, where possible, reconcile instances ofligand-receptor misalignment at sites of peptide action incentral stress-related circuitry in rodent brain (seeHerkenham, 1987; Bittencourt and Sawchenko, 2000). Ac-cordingly, the discussion will focus on novel or controver-

sial aspects of the receptor distributions. A more generalconsideration of the distributions may be found in theoriginal publications on these topics (Potter et al., 1994;Chalmers et al., 1995) and in recent reviews (Behan et al.,1996; Turnbull and Rivier, 1997).

CRF-R1 expression in rodent brain. The distribu-tion of CRF-R1 mRNA in mouse brain was found to befundamentally similar to that described previously in rat(Potter et al., 1994; Chalmers et al., 1995: see also Wong etal., 1994; Rivest et al., 1995; Bonaz and Rivest, 1998), indisplaying predominant localizations in cell groups in-volved in aspects of cortical, cerebellar, and several mo-dalities of sensory information processing, and somewhatmore limited expression in stress-related autonomic andneuroendocrine systems, with which CRF has been moregenerally associated. Although the present methods do notpermit ready comparison across species of the relativestrength of receptor expression in any given cell group,differences in the intensity of expression between cellgroups within species were apparent, and are relevant tothe question of how rats and mice may be used to modelvarious aspects of CRF-R function. Among the regionsdisplaying such differences were: (1) isocortex, where de-spite general similarities in laminar emphases, R1 expres-sion in layer 6 was relatively under-represented in mice,(2) the thalamus, where discrete R1 expression in thereticular nucleus in mouse contrasted sharply with low-level diffuse expression seen in certain ventral, lateral,and midline nuclear groups in rat, and (3) the striatum,where the decidedly patchy distribution of R1 expressionseen in mouse contrasted with the diffuse localization

Fig. 4. Stress-induced corticotropin-releasing factor type 1 recep-tor (CRF-R1) mRNA expression in the paraventricular nucleus of thehypothalamus. Positions of portions of the parvicellular (mp) andmagnocellular (pm) divisions are indicated. Freely moving rats bear-ing jugular catheters were subjected to 15% hemorrhage and killed at

the intervals indicated. As reported by several groups, stress-inducible CRF-R1 expression is seen in both the parvi- and magno-cellular divisions of the nucleus, peaking at 4–6 hours after thechallenge. Scale bar 5 250 mm.


encountered in rat. We are unaware of any pronouncedspecies differences in the connectivity of the patch andmatrix compartments of the neostriatum, and our findingswould therefore imply distinct roles in rat and mouse forCRF-R1 in this structure.

Inducible CRF-R1 expression. Inducible expressionof CRF-R1 mRNA in the paraventricular and/or supraop-tic nuclei has been described by several groups in responseto various stress paradigms (Luo et al., 1994; Rivest et al.,1995; Makino et al., 1995, 1997; Kiss et al., 1996; Lee andRivier, 1997) or central administration of CRF (Mansi etal., 1996). We have extended this to include additionalexamples of so-called physiological (hemorrhage) andemotional (footshock) paradigms, finding induction in thehypophysiotropic zone of the parvicellular division of thePVH in both instances, with responses of the magnocellu-lar compartment and supraoptic nucleus seen preferen-tially following hemorrhage. The CRF-R1 mRNA re-sponses to both acute stresses were remarkable inshowing a slow onset, relative, for example, to stress-induced upregulation of CRF mRNA in the same para-digms. This suggests that it is unlikely that the capacity to

manifest induced receptor expression is involved inmounting adaptive responses during an acute stress.Whether, and under what conditions, the capacity to dis-play inducible CRF-R1 mRNA expression in the PVH maybe manifest in the form of receptor protein expression,remains to be determined. The single study to have suc-cessfully localized CRF-R1-ir in rodent brain to date de-scribes low level basal expression in what appears to cor-respond clearly to the magnocellular, but not theparvicellular, division of the PVH, as well as in the su-praoptic nucleus (Radulovic et al., 1998).

In line with previous observations (Luo et al., 1994;Makino et al., 1995, 1997; Lee and Rivier, 1997), we ad-duced no evidence for any additional sites of CRF-R1induction in either stress model, despite robust activa-tional responses seen in cell groups such as the locuscoeruleus and central nucleus of the amygdala, whichhave been identified as sites of CRF action, but fail todisplay evidence of CRF-R expression. Although severalgroups have described modulatory effects of stress onCRF-R expression at various loci (e.g., Giardino et al.,1996; Makino et al., 1998), the PVH and (more variably)

Fig. 5. Absence of stress-induced alterations in corticotropin-releasing factor type 1 receptor (CRF-R1) mRNA distribution in thelocus coeruleus. Brightfield photomicrographs show immunoperoxi-dase localization of Fos-immunoreactivity (ir; top) and darkfield im-ages of CRF-R1 expression (bottom) in sections through the locuscoeruleus of a control rat (left) and one subjected to acute footshock

stress (right). Despite the fact that acute exposure to electric foot-shock strongly induces expression of the immediate-early gene prod-uct, Fos, in the locus coeruleus (top), no alterations in CRF-RA mRNAin this cell group are evident (bottom). Experimental animal waskilled 3 hours following exposure to a single 30-minute footshocksession. Scale bar 5 250 mm.


Fig. 6. A–F: Corticotropin-releasing factor type 2 receptor (CRF-R2) mRNA distribution in the rat brain. Film autoradiograms show-ing regions displaying positive hybridization signals for CRF-R2mRNA (black), by using a hydrolyzed mouse probe. The distribution ofthis receptor transcript in rat is similar to that illustrated above formouse, although some considerable regional differences in the

strength of hybridization signal between species is evident. Althoughcomparably prominent CRF-R2 localization is seen in the lateralseptum and amygdala of both species, the relative strength of CRF-R2hybridization signal in the bed nucleus, preoptic region, medial hypo-thalamus, and amygdala differ considerably between species. Scalebar 5 2 mm.

the supraoptic nucleus remain the only loci at which in-ducible R1 mRNA expression has been described.

CRF-R2 expression in rodent brain. In agreementwith previous studies (Chalmers et al., 1995; see alsoEghbal-Ahmadi et al., 1998, 1999), the principal loci ofCRF-R2 expression in both species were found to include

the lateral septal, ventromedial hypothalamic, and mes-encephalic raphe nuclei, along with aspects of the amyg-dala and bed nucleus. Most of these show limited capacityfor CRF-R1 expression. Partially overlapping distribu-tions of the two receptor types are seen in restricted as-pects of the amygdala, olfactory system, hippocampal for-

Fig. 7. A–I: Corticotropin-releasing factor type 2 receptor (CRF-R2) mRNA in mouse brain. A series of darkfield photomicrographsarranged from rostral (A) to caudal (I) shows the regional distributionof neurons positively hybridized with an antisense probe generatedfrom a mouse CRF-R2 cDNA isolated from heart. Because CRF-R2mRNA in brain is a smaller variant of the species found in heart,probes were hydrolyzed to Z250-bp fragments for use in brain. Major

sites of expression include aspects of the lateral septum (LS), bednucleus of the stria terminalis (BST), amygdala, the medial preopticand hypothalamic region, the mesencephalic raphe nuclei (DR, MR)and some retrohippocampal cortical fields. Note also that this receptoris expressed in the nucleus of the solitary tract (NTS), one of the areasof CRF/CRF-R1 misalignment. Scale bar 5 2 mm.


mation, hypothalamus and brainstem; possible colo-calization has yet to be addressed experimentally. In ad-dition, our findings inform several controversial aspects ofthe R2 distribution, document some novel localizations,and apparent differences in the emphasis of expressionbetween species speak to functional issues regarding sig-naling by CRF and related ligands in brain.

The available literature on the R2 distribution in ratbrain has failed to detect any substantial expression inisocortical areas (e.g., Chalmers et al., 1995), whereasboth receptor expression and cognate ligand binding hasrecently been described in primate cortex, which is none-theless muted, relative to CRF-R1 expression (MarSanchez et al., 1999). The present findings of CRF-R2expression that predominates in deeper layers of rodentisocortex help reconcile this apparent between-species dis-parity. Our findings of particularly enriched R2 mRNAexpression in aspects of the subicular complex and ento-rhinal cortex appears to represent a departure from thedistribution of this receptor reported in primate. In vitroautoradiographic evidence is available to support the ex-istence of CRF-R2 binding sites in rat entorhinal cortex(Primus et al., 1997).

Conflicting information also exists concerning whetheror not CRF-R2 is expressed in the PVH (cf. Chalmers etal., 1995; Lee and Rivier, 1997; Makino et al., 1998; Mansiet al., 1998), which contains multiple populations ofstress-related visceromotor neurons, including parvicellu-lar neurosecretory CRF-expressing cells that comprise thecentral limb of the hypothalamo-pituitary-adrenal (HPA)axis. In the rat, in which the compartmental organizationof the PVH is more readily discerned (Swanson andSawchenko, 1983), our observations support a relativelylow level of R2 mRNA expression localized discretely tothe magnocellular division of the nucleus. Comparablerelative levels of expression were seen in mouse, and overthe supraoptic nuclei in both species. These findings fail toprovide direct support for substantial R2 mRNA expres-sion by parvicellular neurosecretory neurons.

Although the basic patterns of CRF-R2 mRNA expres-sion in rat and mouse were found to be quite compatible,differences in relative strength of expression within cer-tain cell groups were noted, and are relevant to functionalissues. For example, the starkly preferential distributionof CRF-R1 and -R2 mRNAs in the medial and lateralseptal nuclei in rat has provided a useful framework for

Fig. 8. Corticoptropin-releasing factor receptor (CRF-R) expres-sion in rat hypothalamus and medulla. Darkfield photomicrographsshow the distribution of CRF-R2 mRNA expression in the paraven-tricular (A) and supraoptic (B) nuclei of the hypothalamus, andCRF-R1 (C) and R2 (D) expression in the nucleus of the solitary tract.In the hypothalamus, a low level R2 hybridization signal is seen over

the magnocellular division of the paraventricular nucleus (pm) andover the supraoptic nucleus. In the nucleus of the solitary tract,CRF-R1 expression is limited mainly to the central subnucleus(NTSc), whereas R2 expression is more prominent, and localized tothe dorsal subnucleus (NTSd) and adjoining aspects of the area pos-trema (AP). Scale bars 5 250 mm.


evaluating the specificity with which ligand-containinginputs may target the two receptor systems, and withwhich central peptide injections may impact R1- versusR2-expressing populations (Vaughan et al., 1995; Bitten-court et al., 1999; Bittencourt and Sawchenko, 2000). Thiscompartmental segregation of CRF-R expression appearsto be much less pronounced in mouse. Although we alsofind the murine lateral nucleus to be a dominant site ofCRF-R2 expression, this transcript is expressed moreprominently in the medial septal nucleus of mouse than inrat (see Fig. 2). Coupled with differences in the relativeprominence of CRF-R1 expression between the two spe-cies, the mouse septal region seems clearly less well suitedto model R1 versus R2 targeting than is its counterpartin rat.

A novel localization that holds some potential forreconciling instances of ligand-receptor “mismatch” incentral sites of stress-related CRF peptide action (seebelow) was that of moderate CRF-R2 expression in thenucleus of the solitary tract and adjoining aspects of the

area postrema in both rats and mice. The aspects of thenucleus of the solitary tract (NTS) to which receptortranscripts were localized (dorsal subnucleus) is knownto receive primary baroreceptor afferent input form thecarotid sinus and aortic depressor nerves (Ciriello,1983; Housley et al., 1987), offering some promise fordefining substrates for documented effects of intracere-broventricular and local microinjection of CRF-relatedpeptides on cardiovascular parameters (Brown andFisher, 1985; Brown, 1986).

In view of its historical association with the regulationof food intake, the robust expression of CRF-R2 in theventromedial nucleus of the hypothalamus in rat has at-tracted attention as a potential site of action in mediatingthe suppression of appetite seen following central admin-istration of CRF or UCN (Spina et al., 1996; Makino et al.,1998; Smagin et al., 1998; Ohata et al., 2000). Althoughstill prominent, expression of R2 mRNA in the mouseventromedial nucleus is, again on a relative basis, no moreso than in other aspects of the hypothalamus that have

Fig. 9. Partial overlap in sites of corticoptropin-releasing factortype 2 receptor (CRF-R2) and androgen receptor expression. Bright-field photomicrographs of rat brain sections prepared for concurrentimmunoperoxidase localization of androgen receptor (AR) immunore-activity (brown nuclear label) and hybridization histochemical local-ization of CRF-R2 mRNA (black silver grains). In the encapsulatedpart of the bed nucleus of the stria terminalis (BSTe), and the poste-rior part of the medial nucleus of the amygdala (MeAp), diffuse R2

mRNA signal codistributes with AR-ir neurons. In the intermediatelateral septal nucleus (LSi) and the dorsomedial aspect of the ventro-medial hypothalamic nucleus (VMH), the distributions of R2 mRNAand AR-immunoreactivity (ir) are largely distinct, although examplesof doubly labeled cells (arrowheads) are seen in regions of partialoverlap. Filled and open arrows identify cells labeled only for AR-ir orCRF-R2 mRNA, respectively. Scale bar 5 50mm.


also been implicated in the control of feeding and energybalance, such as the arcuate nucleus (lateral aspect) andlateral hypothalamic area (e.g., Sawchenko, 1998). Addingto the complexity is our observation of CRF-R1 expressionin topographically distinct (medial) portions of the arcuatenucleus.

The distribution of CRF-R2 in the limbic forebrain, inparticular, displayed marked similarities with those re-ported for the gonadal steroid hormone receptors, partic-ularly the androgen receptor (Simerly et al., 1990). Exper-iments involving dual localization of androgen receptor-irand R2 transcripts revealed substantial congruence in theencapsulated part of the bed nucleus of the stria termina-lis and the medial amygdaloid nuclei, with somewhatlesser apparent overlap in the medial preoptic, ventrome-dial hypothalamic, and lateral septal nuclei. Owing to thecompromise in the sensitivity of constituent methods insuch combined applications, and the relatively low reso-lution of isotopic hybridization histochemistry, we suspectthat our descriptions underestimate the actual degree ofoverlap. These cell groups have been identified as compris-ing a highly interconnected set of structures that exhibitsteroid-dependent sexual dimorphisms in cellular archi-tecture and/or neurochemical phenotype, and play prom-inent roles in reproductive physiology and behavior (Si-merly, 1990). This provides a basis for evaluating the roleof CRF-R2 mechanisms in a novel context.

CRF-R expression in pituitary. In rat pituitary,CRF-R1 was found previously to be expressed in the in-termediate lobe, and in a subset of anterior lobe cells thatconform mainly to corticotropes (Potter et al., 1994; seealso Chalmers et al., 1995). This defined a basis for theacknowledged principal action of CRF in initiatingpituitary-adrenal responses to stress. The situation asregards the type 2 receptor is more unsettled, with rela-tively weak R2 mRNA signal having been described overscattered cells in rat anterior lobe (Chalmers et al., 1995),whereas more substantial R2 mRNA expression and li-gand binding have been noted in primate adenohypophy-sis (Mar Sanchez et al., 1999). Our results are equivocalconcerning expression in the anterior lobe, as a diffuse

low-level above-control signal lacked cellular resolution,but we did find support for a substantial CRF-R2 presencein the posterior lobe of the gland. Modified astroglia, orpituicytes, that comprise the dominant cell type of theposterior lobe, are the most likely seats of R2 expression atthis locus. By virtue of their capacity to dynamically ex-tend and retract processes as a lawful function of demandon the magnocellular neurosecretory system, pituicyteshave been implicated as playing a role in regulating accessof oxytocin and vasopressin to the perivascular space, andhence the general circulation (see Hatton, 1990, for areview). R2 expressed in the posterior pituitary wouldseem in a position to function in this context.

Implications for signaling by CRF-related ligands in

stress circuits. In addition to its acknowledged role ininitiating pituitary adrenal responses to stress, CRF iswidely distributed in brain and can act centrally to elicitsuch stress-like effects as activation of the sympathoa-drenal system (Brown and Fisher, 1985; see Fisher,1993), generalized arousal and anxiety-like behaviors(Sutton et al., 1982; see Koob et al., 1993), suppressionof immune functions (Irwin et al., 1988) and appetitivebehavior (Gosnell et al., 1983; Spina et al., 1996). Suchobservations gave rise to the hypothesis that CRF mayserve as an integrator of multiple components of thewhole animal response to stress, an idea which hasdominated this field of study since the discovery of thepeptide. Most central sites of CRF action in elicitingstress-like behavioral and autonomic responses conformto a highly interconnected set of cell groups known to bepivotally involved in the processing of visceral sensoryinformation (Sawchenko, 1983; Saper, 1995). Includedamong this grouping are central nucleus of the amyg-dala and associated parts of the bed nucleus of the striaterminalis (Brown and Fisher, 1985; Koob et al., 1993),the PVH (Brown, 1986; Krahn et al., 1988; Parkes et al.,1993), the lateral parabrachial nucleus, and the nucleusof the solitary tract and ventrolateral medulla (Brown,1986; Fisher, 1993; Milner et al., 1993). Also worthy ofconsideration in this regard is the locus coeruleus,whose widespread noradrenergic output is thought to

Fig. 10. Corticotropin-releasing factory receptor (CRF-R) distributions in rodent brain. Schematicdrawing of a sagittal section through the rat brain shows the distribution and relative density of cellsexpressing CRF-R1 and CRF-R2 mRNAs. The CRF-R2 transcript shows a more restricted distributionthat is largely nonoverlapping with that of CRF-R1.


participate in setting levels of arousal and “behavioralvigilance” (Foote et al., 1983), and where unit activity isenhanced by central or local administration of CRF(Valentino, 1990). These cell groups are quite uniformlyrecruited to activation in response to central adminis-tration of CRF or UCN (e.g., Bittencourt and Sawchenko,2000), as well as in response to a wide variety of stressparadigms (review: Sawchenko et al., 1996).

A stern challenge to the hypothesized role of CRF as anintegrator of multiple modalities of stress responses isposed by the fact that although most of the cell groupsenumerated above have been identified as sites of CRFaction in eliciting stress-like responses, they display verylimited capacity for CRF-R expression (e.g., Potter et al.,1994; Chalmers et al., 1995) or CRF binding (De Souza etal., 1985). Until now, only in the lateral parabrachialnucleus, which receives a CRF-ir input (Swanson et al.,1983) and expresses CRF-R1 (Potter et al., 1994), was anysubstantial degree of ligand-receptor alignment apparentin this system. The present observations of CRF-R2 ex-pression in the nucleus of the solitary tract offers potentialfor further reconciliation of such mismatches, particularlyin view of the recent demonstration of UCN-ir inputs tothis very region (Bittencourt et al., 1999). Other majorstress-related sites of CRF peptide action, particularly inthe locus coeruleus and central nucleus of the amygdala,have thus far failed to display a capacity for CRF-R ex-pression. These disparities may ultimately be accommo-dated by identification of still additional CRF receptorsubtypes, or perhaps by situational induction of receptorexpression, as has been shown for the CRF-R1 in theparaventricular nucleus.

ACKNOWLEDGMENTS

We are grateful to Neurocrine Biosciences for gener-ously sharing CRF-R2 probes and unpublished data, andto Belle Wamsley and Kris Trulock for excellent editorialand photography/graphics assistance, respectively. Grantsponsor: NIH; grant number: DK-26741 and in part by theFoundation for Research and the Foundation for MedicalResearch. Fellowship support: Medical Research Councilof Canada (V.V.). FAPESP grant 99/00213-5 (J.C.B.) andthe American Heart Association-California Affiliate, grantnumber 95-117 (H.-Y.L.).

LITERATURE CITED

Asaba K, Makino S, Hashimoto K. 1998. Effect of urocortin on ACTHsecretion from rat anterior pituitary in vitro and in vivo: comparisonwith corticotropin-releasing hormone. Brain Res 806:95–103.

Bale TL, Smith GW, Contarino A, Chan R, Gold L, Sawchenko PE, KoobGF, Vale WW, Lee K-F. 2000. Corticotropin releasing factor receptor2-deficient mice display anxiety-like behavior and are hypersensitive tostress. Nat Genet 24:410–414.

Behan DP, Grigoriadis DE, Lovenberg T, Chalmers D, Heinrichs S, Liaw C,De Souza EB. 1996. Neurobiology of corticotropin releasing factor(CRF) receptors and CRF-binding protein: implications for the treat-ment of CNS disorders. Mol Psych 1:265–277.

Bittencourt JC, Sawchenko PE. 2000. Do centrally administered neuropep-tides access cognate receptors?: an analysis in the central corticotropin-releasing factor system. J Neurosci 20:1142–1156.

Bittencourt JC, Vaughan J, Arias C, Vale WW, Sawchenko PE. 1999.Urocortin expression in rat brain: evidence against a pervasive rela-tionship of urocortin-containing projections with targets bearing type 2CRF receptors. J Comp Neurol 415:285–312.

Bonaz B, Rivest S. 1998. Effect of a chronic stress on CRF neuronal activity

and expression of its type 1 receptor in the rat brain. Am J Physiol275:R1438–R1449.

Brown M. 1986. Corticotropin releasing factor: central nervous systemsites of action. Brain Res 399:10–14.

Brown MR, Fisher LA. 1985. Corticotropin-releasing factor: effects on theautonomic nervous system and visceral system. Fedn Proc 44:243–248.

Chadwick DJ, Marsh J, Ackrill K, editors. 1993. Corticotropin-releasingfactor. Ciba Foundation Symposium 172. London: John Wiley & Sons.

Chalmers DT, Lovenberg TW, De Souza EB. 1995. Localization of novelcorticotropin-releasing factor receptor (CRF2) mRNA expression tospecific sub-cortical nuclei in rat brain: comparison with CRF1 receptormRNA expression. J Neurosci 15:6340–6350.

Chan RKW, Sawchenko PE. 1994. Spatially and temporally differentiatedpatterns of c-fos expression in brainstem catecholaminergic cell groupsinduced by cardiovascular challenges in the rat. J Comp Neurol 348:433–460.

Chan RKW, Brown ER, Ericsson A, Kovacs KJ, Sawchenko PE. 1993. Acomparison of two immediate-early genes, c-fos and NGFI-B, as mark-ers for functional activation in stress-related neuroendocrine circuitry.J Neurosci 13:5125–5138.

Chang CP, Pearse RV II, O’Connell S, Rosenfeld MG. 1993. Identificationof a seven transmembrane helix receptor for corticotropin-releasingfactor and sauvigine in mammalian brain. Neuron 11:1187–1195.

Chen R, Lewis KA, Perrin MH, Vale WW. 1993. Expression cloning of ahuman corticotropin-releasing-factor receptor. Proc Natl Acad Sci USA90:8967–8971.

Ciriello J. 1983. Brainstem projections of aortic baroreceptor afferent fibersin the rat. Neurosci Lett 36:37–42.

Coste SC, Kesterson RA, Heldwein KA, Stevens SL, Heard AD, Hollis JH,Murray, SE, Hill, JK, Pantley GA, Hohimer AR, Hatton DC, PhillipsTJ, Finn DA, Rittenberg MB, Stenzel P, Stenzel-Poore MP. 2000.Abnormal adaptations to stress and impaired cardiovascular functionin mice lacking corticotropin-releasing hormone receptor-2. Nat Genet24:403–409.

Cox KH, DeLeon DV, Angerer LM, Angerer RC. 1984. Detection of mRNAsin sea urchin embryos by in situ hybridization using asymmetric RNAprobes. Dev Biol 101:485–502.

Cunningham ET, Jr, Sawchenko PE. 1989. A circumscribed projection fromthe nucleus of the solitary tract to the nucleus ambiguus in the rat:anatomical evidence for somatostatin-28-immunoreactive interneuronssubserving reflex control of esophageal motility. J Neurosci 9:1668–1682.

De Souza EB, Insel TR, Perrin MH, Rivier J, Vale WW, Kuhar MJ. 1985.Corticotropin-releasing factor receptors are widely distributed in therat central nervous system: an autoradiographic study. J Neurosci5:3189–3203.

Eghbal-Ahmadi M, Hatalski CG, Lovenberg TW, Avishai-Eliner S, Chalm-ers DT, Baram TZ. 1998. The developmental profile of the corticotropinreleasing factor receptor (CRF2) in rat brain predicts distinct age-specific functions. Brain Res Dev Brain Res 107:81–90.

Eghbal-Ahmadi M, Avishai-Eliner S, Hatalski CG, Baram TZ. 1999. Dif-ferential regulation of the expression of corticotropin-releasing factorreceptor type 2 (CRF2) in hypothalamus and amygdala of the imma-ture rat by sensory input and food intake. J Neurosci 19:3982–3991.

Fisher LA. 1993. Central actions of corticotropin-releasing factor on auto-nomic nervous activity and cardio-vascular functioning. In: ChadwickDJ, Marsh J, Ackrill, K, editors. Corticotropin-releasing factor. CibaFoundation Symposium 172. London: John Wiley & Sons. p 243–257.

Foote SL, Bloom FE, Aston-Jones G. 1983. Nucleus locus ceruleus: newevidence of anatomical and physiological specificity. Physiol Rev 63:844–914.

Fulwiler CE, Saper CB. 1984. Subnuclear organization of the efferentconnections of the parabrachial nucleus in the rat. Brain Res Rev7:229–259.

Giardino L, Puglisi-Allegra S, Ceccatelli S. 1996. CRH-R1 mRNA expres-sion in tow strains of inbred mice and its regulation after repeatedrestraint stress. Brain Res Mol Brain Res 40:310–314.

Gosnell BA, Morley JE, Levine AS. 1983. Adrenal modulation of the inhib-itory effect of corticotropin releasing factor on feeding. Peptides 4:807–812.

Hatton GI. 1990. Emerging concepts of structure-function dynamics inadult brain: the hypothalamo-neurohypophysial system. Prog Neuro-biol 34:437–504.


Herkenham M. 1987. Mismatches between neurotransmitter and receptorlocalizations in brain: observations and implications. Neuroscience 23:1–38.

Housley GD, Martin-Body RL, Dawson NJ, Sinclair JD. 1987. Brain stemprojections of the glossopharyngeal nerve and its carotid sinus branchin the rat. Neuroscience 22:237–250.

Irwin M, Hauger RL, Brown M, Britton KT. 1988. CRF activates autonomicnervous system and reduces natural killer cytotoxicity. Am J Physiol255:R744–R747.

Ju G, Swanson LW. 1989. Studies on the cellular architecture of the bednuclei of the stria terminalis in the rat: I. Cytoarchitecture. J CompNeurol 280:587–602.

Karolyi IJ, Burrows HL, Ramesh TM, Nakajima M, Lesh JS, Seong E,Camper SA, Seasholtz AF. 1999. Altered anxiety and weight gain incorticotropin-releasing hormone-binding protein-deficient mice. ProcNatl Acad Sci USA 96:11595–11600.

Kishimoto T, Radulovic J, Radulovic M, Lin CR, Schrick C, Hooshmand F,Hermanson O, Rosenfeld MG, Spiess J. 2000. Deletion of crhr2 revealsan anxiolytic role for corticotropin-releasing hormone receptor-2. NatGenet 24:415–419.

Kiss A, Palkovits M, Aguilera G. 1996. Neural regulation of corticotropinreleasing hormone (CRH) and CRH receptor mRNA in the hypotha-lamic paraventricular nucleus in the rat. J Neuroendocrinol 8:103–112.

Koob GF, Heinrichs SC, Pich EM, Menzaghi F, Baldwin H, Miczek K,Britton KT. 1993. The role of CRF in behavioral responses to stress. In:Widdows K, editor. Corticotropin-releasing factor. Ciba FoundationSymposium No. 172. London: John Wiley & Sons. p 277–295.

Kozicz T, Yanaihara H, Arimura A. 1998. Distribution of urocortin-likeimmunoreactivity in the central nervous system of the rat. J CompNeurol 391:1–10.

Krahn DD, Gosnell BA, Levine AS, Morley JE. 1988. Behavioral effects ofcorticotropin-releasing factor: localization and characterization of cen-tral effects. Brain Res 443:63–69.

Lee S, Rivier C. 1997. Alcohol increases the expression of type 1, but nottype 2 a corticotropin-releasing factor (CRF) receptor messenger ribo-nucleic acid in the rat hypothalamus. Brain Res Mol Brain Res 52:78–89.

Li H-Y, Sawchenko PE. 1998. Hypothalamic effector neurons and extendedcircuitries activated in “neurogenic” stress: a comparison of footshockeffects exerted acutely, chronically, and in animals with controlledglucocorticoid levels. J Comp Neurol 393:244–266.

Lovenberg TW, Liaw CW, Grigoriadis DE, Clevenger W, Chalmers DT, DeSouza EB, Oltersdorf T. 1995a. Cloning and characterization of afunctionally distinct corticotropin-releasing factor receptor subtypefrom rat brain. Proc Natl Acad Sci USA 92:836–840.

Lovenberg TW, Chalmers DT, Liu C, De Souza EB. 1995b. CRF2a

andCRF

2breceptor mRNAs are differentially distributed between the rat

central nervous system and peripheral tissues. Endocrinology 136:4139–4142.

Luo X, Kiss A, Makara G, Lolait SJ, Aguilera G. 1994. Stress-specificregulation of corticotropin releasing hormone receptor expression inthe paraventricular and supraoptic nuclei of the hypothalamus in therat. J Neuroendocrinol 6:689–696.

Makino S, Schulkin J, Smith MA, Pacak K, Palkovits M, Gold PW. 1995.Regulation of corticotropin-releasing hormone receptor messenger ri-bonucleic acid in the rat brain and pituitary by glucocorticoids andstress. Endocrinology 136:4517–4525.

Makino S, Takemura T, Asaba K, Nishiyama M, Takao T, Hashimoto K.1997. Differential regulation of type-1 and type-2a corticotropin-releasing hormone receptor mRNA in the hypothalamic paraventricu-lar nucleus of the rat. Brain Res Mol Brain Res 47:170–176.

Makino S, Nishiyama M, Asaba K, Gold PW, Hashimoto K. 1998. Alteredexpression of type 2 CRH receptor mRNA in the VMH by glucocorti-coids and starvation. Am J Physiol 275:R1138–R1145.

Mansi JA, Rivest S, Drolet G. 1996. Regulation of corticotropin-releasingfactor type 1 (CRF1) receptor messenger ribonucleic acid in the para-ventricular nucleus of rat hypothalamus by exogenous CRF. Endocri-nology 137:4619–4629.

Mansi JA, Rivest S, Drolet G. 1998. Effect of immobilization stress ontranscriptional activity of inducible immediate-early genes,corticotropin-releasing factor, its type 2 receptor, and enkephalin in thehypothalamus of borderline hypertensive rats. J Neurochem 70:1556–1566.

Mar Sanchez M, Young LJ, Plotsky PM, Insel TR. 1999. Autoradiographicand in situ hybridization localization of corticotropin-releasing factor 1

and 2 receptors in nonhuman primate brain. J Comp Neurol 408:365–377.

Merchenthaler I, Vigh S, Petrusz P, Schally AV. 1982. Immunocytochem-ical localization of corticotropin-releasing factor (CRF) in the rat brain.Am J Anat 165:385–396.

Milner TA, Reis DJ, Pickel VM, Aicher SA, Giuliano R. 1993. Ultrastruc-tural localization and afferent sources of corticotropin-releasing factorin the rat rostral ventrolateral medulla: implications for central car-diovascular regulation. J Comp Neurol 333:151–167.

Muglia L, Jacobson L, Dikkes P, Majzoub JA. 1995. Corticotropin-releasinghormone deficiency reveals major fetal but not adult glucocorticoidneed. Nature 373:427–432.

Ohata H, Suzuki K, Oki Y, Shibasaki T. 2000. Urocortin in the ventrome-dial hypothalamic nucleus acts as an inhibitor of feeding behavior inrats. Brain Res 861:1–7.

Parkes D, Rivest S, Lee S, Rivier C, Vale W. 1993. Corticotropin-releasingfactor activates c-fos, NGFI-B, and corticotropin-releasing factor geneexpression within the paraventricular nucleus of the rat hypothala-mus. Mol Endocrinol 7:1357–1367.

Perrin M, Donaldson C, Chen R, Blount A, Berggren T, Bilezikjian L,Sawchenko PE, Vale W. 1995. Identification of a second CRF receptorgene and characterization of a cDNA expressed in heart. Proc NatlAcad Sci USA 92:2969–2973.

Potter E, Sutton S, Donaldson C, Chen R, Perrin M, Lewis K, SawchenkoPE, Vale W. 1994. Distribution of corticotropin-releasing factor recep-tor mRNA expression in the rat brain and pituitary. Proc Natl Acad SciUSA 91:8777–8781.

Primus RJ, Yevich E, Baltazar C, Gallager DW. 1997. Autoradiographiclocalization of CRF1 and CRF2 binding sites in adult rat brain. Neu-ropsychopharmacology 17:308–316.

Prins GS, Birch L, Greene GL. 1991. Androgen receptor localization indifferent cell types of the adult rat prostate. Endocrinology 129:3187–3199.

Radulovic J, Sydow S, Spiess J. 1998. Characterization of nativecorticotropin-releasing factor receptor type 1 (CRFR1) in the rat andmouse central nervous system. J Neurosci Res 54:507–521.

Rivest S, Laflamme N, Nappi RE. 1995. Immune challenge and immobili-zation stress induce transcription of the gene encoding the CRF recep-tor in selective nuclei of the rat hypothalamus. J Neurosci 15:2680–2695.

Ross CA, Ruggiero DA, Reis DJ. 1985. Projections from the nucleus tractussolitarii to the rostral ventrolateral medulla. J Comp Neurol 242:511–534.

Sakanaka M, Shibasaki T, Lederis K. 1987. Corticotropin-releasing factor-like immunoreactivity in the rat brain as revealed by a modified cobalt-glucose oxidase-diaminobenzidine method. J Comp Neurol 260:256–298.

Saper CB. 1995. Central autonomic system. In: Paxinos G, editor. The ratnervous system, 2nd edition. San Diego: Academic Press. p 107–128.

Sawchenko PE. 1983. Central connections of the sensory and motor nucleiof the vagus nerve. J Auton Nerv Syst 9:13–26.

Sawchenko PE. 1998. Toward a new neurobiology of energy balance, ap-petite, and obesity: the anatomists weigh in. J Comp Neurol 402:435–441.

Sawchenko PE, Swanson LW. 1989. Organization of CRF immunoreactivecells and fibers in the rat brain: immunohistochemical studies. In: DeSouza EB, Nemeroff CB, editors. Corticotropin-releasing factor: basicand clinical studies of a neuropeptide. Boca Raton, FL: CRC Press.p 29–51.

Sawchenko PE, Cunningham ET, Jr, Mortrud MT, Pfeiffer SW, Gerfen CR.1990. Phaseolus vulgaris -leucoagglutinin (PHA-L) anterograde axonaltransport technique. In: Conn PM, editor. Methods in neurosciences.New York: Academic Press. p 247–260.

Sawchenko PE, Imaki T, Potter E, Kovacs K, Imaki J, Vale W. 1993. Thefunctional neuroanatomy of corticotropin-releasing factor. In: Chad-wick DJ, Marsh J, Ackrill K, editors. Corticotropin-releasing factor.Ciba Foundation Symposium 172. London: John Wiley & Sons. p 5–29.

Sawchenko PE, Brown ER, Chan RKW, Ericsson A, Li H-Y, Roland BL,Kovacs KJ. 1996. The paraventricular nucleus of the hypothalamusand the functional neuroanatomy of visceromotor responses to stress.In: Holstege G, Bandler R, Saper CB, editors. Progress in brain re-search, the emotional motor system. Amsterdam: Elsevier. p 201–222.

Simerly RB. 1990. Hormonal control of neuropeptide gene expression insexually dimorphic olfactory pathways. Trends Neurosci 13:104–110.


Simerly RB, Chang C, Muramatsu M, Swanson LW. 1990. The distri-bution of androgen and estrogen receptor mRNA-containing cells inthe rat brain: an in situ hybridization study. J Comp Neurol 294:76 –95.

Simmons DM, Arriza JL, Swanson LW. 1989. A complete protocol for insitu hybridization of messenger RNAs in brain and other tissues withradiolabeled single-stranded RNA probes. J Histotechnol 12:169–181.

Smagin GN, Howell LA, Ryan DH, De Souza EB, Harris RB. 1998. The roleof CRF2 receptors in corticotropin-releasing factor- and urocortin-induced anorexia. Neuroreport 9:1601–1606.

Smith GW, Aubry J-M, Dellu F, Contarino A, Bilezikjian LM, Gold LH,Chen R, Marchuk Y, Hauser C, Bentley CA, Sawchenko PE, Koob GF,Vale W, Lee K-F. 1998. Corticotropin releasing factor receptor1-deficient mice display decreased anxiety, impaired stress response,and aberrant neuroendocrine development. Neuron 20:1093–1102.

Spina M, Merlo-Pich E, Chan RKW, Basso AM, Rivier J, Vale W, Koob GF.1996. Appetite-suppressing effects of urocortin, a CRF-related neu-ropeptide. Science 273:1561–1564.

Sutton RE, Koob GF, LeMoal M, Rivier J, Vale W. 1982. Corticotropin-releasing factor (CRF) produces behavioral activation in rats. Nature297:331–333.

Swanson LW, Sawchenko PE. 1983. Hypothalamic integration: organiza-tion of the paraventricular and supraoptic nuclei. Ann Rev Neurosci6:269–324.

Swanson LW, Sawchenko PE, Rivier J, Vale W. 1983. Organization of ovinecorticotropin-releasing factor immunoreactive cells and fibers in the ratbrain: an immunohistochemical study. Neuroendocrinology 36:165–186.

Turnbull AV, Rivier C. 1997. Corticotropin-releasing factor (DRF) andendocrine responses to stress: CRF receptors, binding protein, andrelated peptides. Proc Soc Exp Biol Med 215:1–10.

Valentino RJ. 1990. Effects of CRF on spontaneous and sensory-evokedactivity of locus coeruleus neurons. In: De Souza EB, Nemeroff CB,editors. Corticotropin releasing factor: basic and clinical studies of aneuropeptide. Boca Raton, FL: CRC Press. p 217–231.

Vaughan J, Donaldson C, Bittencourt J, Perrin MH, Lewis K, Sutton S,Chan R, Turnbull AV, Lovejoy D, Rivier C, Rivier J, Sawchenko PE,Vale W. 1995. Urocortin, a mammalian neuropeptide related to fishurotensin I and to corticotropin-releasing factor. Nature 378:287–292.

Vita N, Laurent P, Lefort S, Chalon P, Lelias J-M, Kaghad M, Le Fur G,Caput D, Ferrara P. 1993. Primary structure and functional expressionof mouse pituitary and human brain corticotrophin releasing factorreceptors. FEBS Lett 335:1–5.

Wong M-L, Licinio J, Pasternak KI, Gold PW. 1994. Localization ofcorticotropin-releasing hormone (CRH) receptor mRNA in adult ratbrain by in situ hybridization histochemistry. Endocrinology 135:2275–2278.


Distribution of mRNAs encoding CRF receptors in brain and pituitary of rat and mouse

Documents