Parathyroid hormone 2 receptor and its endogenous ligand tuberoinfundibular peptide of 39 residues are concentrated in endocrine, viscerosensory and auditory brain regions in macaque

iPt

H

HEHMwMmteoocstvobic

opfw4da

M

BssistccytmBmasmsdtu

R

TbmsgeAcwiopapPps23Gs6NpvuopouPal

I

Fatcdtsmact

fScamofb

T

B

12345

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147130

nvestigated the anatomical relationship between theTH2R and hypophysiotropic somatostatin-, and cortico-

ropin-releasing hormone (CRH)–containing neurons.

EXPERIMENTAL PROCEDURES

uman tissue

uman brain samples were collected in accordance with thethical Rules for Using Human Tissues for Medical Research inungary (HM 34/1999) and the Code of Ethics of the Worldedical Association (Declaration of Helsinki). Tissue samplesere taken during brain autopsy at the Department of Forensicedicine of Semmelweis University in the framework of the Hu-an Brain Tissue Bank, Budapest, or at the Department of Pa-

hology of the University of Pécs, as approved by institutionalthics committees of the Semmelweis University or the Universityf Pécs. Prior written informed consent was obtained from the nextf kin, which included the request to consult the medical chart and toonduct neurochemical analyses. The study reported in the manu-cript was performed according to animal care protocols approved byhe Committee of Science and Research Ethics, Semmelweis Uni-ersity (TUKEB 34-1/2002). The medical history of the subjects wasbtained from medical or clinical records, interviews with family mem-ers and relatives, as well as from pathological and neuropatholog-

cal reports. All personal identifiers had been removed and samplesoded before the analyses of tissue.

Brains were removed from the skull with a postmortem delayf 2–6 h. For microdissection, the brains were cut into five largearts (cerebral lobes, diencephalon, brainstem, cerebellum), androzen immediately at �80 °C. For immunocytochemistry, brainsere cut into 5–10 mm thick coronal slices and immersion fixed in% paraformaldehyde in 0.1 M phosphate buffer (PB) for 6–10ays. Then, the slices were postfixed in the same solution withddition of 15% saturated picric acid.

icrodissection of human brain tissue

rain nuclei and areas including the frontal cortex, hippocampus,eptum, caudate nucleus, amygdala, ventral thalamus, mediodor-al thalamic nucleus, pulvinar, lateral and medial geniculate bod-es, subthalamic nucleus, medial hypothalamus, pretectal area,ubstantia nigra, ventral tegmental area, pontine nuclei, tegmen-um and reticular formation, ventrolateral medulla, dorsal vagalomplex, inferior olive, spinal trigeminal nucleus, and cerebellarortex were individually microdissected from the brains of an 89ear old woman and a 56 year old man using the micropunchechnique (Palkovits, 1973; Palkovits et al., 2008) guided by hu-an brain atlases (Paxinos and Huang, 1995; Mai et al., 1997).riefly, the large parts of the frozen brains were cut into 1.0–1.5m thick coronal sections by an electric slicer at about �5–10 °C,nd individual brain regions and nuclei were removed from thelides by special punch needles with an inside diameter of 1.0–3.5m visualized using either a head magnifier, or a stereomicro-

cope. The slices were kept on dry-ice during the whole proce-ure. The microdissected samples were collected in 2.0 mm air-ight plastic (Eppendorf) tubes and stored at �80 °C until furtherse.

able 1. Information on human tissue used in the study

rain number Gender Age (years) Clinical diagnosis

Female 89 Alzheimer diseaseMale 56 Cardiac failureFemale 8 LeukemiaMale 10 Leukemia

T-PCR of the PTH2R from human

otal mRNA was isolated using Trizol reagent (Invitrogen, Carls-ad, CA, USA) from 50 to 100 mg of brain tissue according to theanufacturer’s instructions. The degradation of RNA was as-

essed by running the purified RNAs on denaturing formaldehydeels. Samples in which the amount of 28 S rRNS was at leastqual to that of 18 S rRNA were processed further for RT-PCR.fter diluting total RNA to 2 �g/�l, RNA was treated with amplifi-ation Grade DNase I (Invitrogen) and cDNA was synthesizedith a Superscript II reverse transcriptase kit (Invitrogen) accord-

ng to the manufacturer’s instructions. After 10-fold dilution, 2.5 �lf the resulting cDNA was used as template in PCR reactions. Therimer pair for PTH2R was 5=-CAATTGCTTGGCTGTAGCTTT-3=nd 5=-ACAAAATCAATTTGCAGACACAA-3= resulting in a PCRroduct of 440 bp that corresponds to bp 2162-2601 of the humanTH2R (GenBank accession number NM_005048). The PCRroduct made using this PTH2R primer pair has been verified byequencing to result in a product specific for the PTH2R (Bago et al.,008). The primer pair for the housekeeping gene glyceraldehyde--phosphate dehydrogenase (GAPDH) was 5=-CCACCCAGAA-ACTGTGGAT-3= and 5=-CCCTGTTGCTGTAGCCAAAT-3= re-ulting in a PCR product of 423 bp that corresponds to bp50-1072 of human GAPDH (GenBank accession numberM_002046). The PCR reactions were performed with iTaq DNAolymerase (Bio-Rad Laboratories, Hercules, CA, USA) in totalolumes of 12.5 �l using primers at 300 nM final concentrationnder the following conditions: 95 °C for 3 min, followed by cyclesf 95 °C for 0.5 min, 60 °C for 0.5 min and 72 °C for 1 min. Theresence the PTH2R was evaluated after 38 cycles; the presencef GAPDH, after 33 cycles. Equal amounts (10 �l) of PCR prod-cts were run on gels and pictures taken with a digital camera.rimers for PTH2R are specific to sequences in different exons tollow recognition of potential genomic DNA contamination by the

arger size of the product.

mmunolabeling of PTH2R in human tissue

or immunocytochemistry, coronally oriented tissue blocks ofbout 10�10�20 mm from three human brains were sectioned inhe coronal plane (Table 1). One tissue block containing insularortex and continuous tissue blocks from the rostral end of theiencephalon (from 1 mm rostral to the anterior commissure) tohe caudal end of the brainstem from a 10 year old male wasectioned. Another tissue block comprising the entire hypothala-us from the brain of an 8 year old female was sectioned. Inddition, two tissue blocks from a 62 year old male containing theaudal part of the medulla oblongata (from the level of the obex)o the rostral part of the cervical spinal cord were also sectioned.

Two days before sectioning, the tissue blocks were trans-erred to PB for 2 days to remove the excess paraformaldehyde.ubsequently, the blocks were placed in PB containing 20% su-rose for 2 days for cryoprotection. Then, the blocks were frozen,nd cut into 50 �m thick serial coronal sections on a slidingicrotome. Immunolabeling was performed as described previ-usly (Dobolyi et al., 2006a) on every 10th section. Briefly, theree-floating sections were pretreated with 3% bovine serum al-umin in PB containing 0.5% Triton X-100 for 30 min at room

ue Detection method

h frozen micropunched samples RT-PCRh frozen micropunched samples RT-PCRersion fixed hypothalamus Immunolablelingersion fixed insula, diencephalon, brainstem Immunolableling

Tiss

FresFresImmImm

ersion fixed medulla and spinal cord (C1, 2) Immunolableling

tmTpeaWmrBctflHmaP

Ds

AmlmssLTalCLAMCCiaaPsetdphTwlpmt52s

M

TpNafifame

Tai

9iaprribs

anTwamdoci

M

Tre(TGqafpqeaBtpaabsdnqD

I

Sbbl(pbdmgtdse

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 131

emperature. The sections were then placed in anti-PTH2R pri-ary antiserum (1:20,000 dilution) for 48 h at room temperature.his antiserum has previously been used and characterized. Itroduced strong specific immunolabeling of HEK293 cells stablyxpressing the human PTH2R and its labeling was blocked bybsorption with the peptide immunogen (Usdin et al., 1999a;ang et al., 2000). Following the incubation in anti-PTH2R pri-ary antiserum, the sections were placed in biotinylated anti-

abbit secondary antibody (1:600 dilution; Vector Laboratories,urlingame, CA, USA) for 2 h followed by incubation in a solutionontaining avidin–biotin–peroxidase complex (1:300 dilution; Vec-or Laboratories) for 2 h. The sections were then treated withuorescein isothiocyanate (FITC)-tyramide (1:8000 dilution) and2O2 (0.003%) in Tris hydrochloride buffer (0.05 M, pH 8.2) for 6in. The sections were then mounted, dried, and coverslipped inntifade mounting medium (Prolong Antifade Kit, Molecularrobes, Eugene, OR, USA).

ouble immunolabeling of PTH2R and CRH oromatostatin in human hypothalamus

second set of hypothalamic sections from the 10-year old hu-an was used for double immunolabeling. First, PTH2R immuno-

abeling was performed as for single labeling except for usingore dilute anti-PTH2R primary antiserum (1:40,000 dilution), as

uggested previously (Hunyady et al., 1996). Subsequently, theections were incubated in rabbit anti-CRH antibody (Peninsulaaboratories Inc., San Carlos, CA, USA; catalogue number-4037, lot number 041048-5; 1:6000) or rabbit anti-somatostatinntibody (Peninsula Laboratories Inc.; catalogue number T-4103,

ot number 010965-7; 1:2000). The antigen sequence of the anti-RH antibody is H–Ser–Glu–Glu–Pro–Pro–Ile–Ser–Leu–Asp–eu–Thr–Phe–His–Leu–Leu–Arg–Gl–Val–Leu–Glu–Met–Ala–Arg–la–Glu–Gln–Leu–Ala–Gln–Gln–Ala–His–Ser–Asn–Arg–Lys–Leu–et–Glu–Ile–Ile–NH2. This antibody recognizes human and ratRH with equal specificity. Its cross-reactivity is 10% to porcineRH, 5% to savagine, and 0.05% to ovine and bovine CRH, and

t does not recognize human prepro-CRH, adrenocorticotropin,nd PACAP. The antigen sequence of the anti-somatostatinntibody is H–Ala–Gly–Cys–Lys–Asn–Phe–Phe–Trp–Lys–Thr–he–Thr–Ser–Cys–OH (disulfide bond). This antibody recognizesomatostatin-14, -28, -25 and (des-Ala12)-somatostatin-14 withqual specificity. Its cross-reactivity is 0.05% to (D-Trp8)-soma-ostatin-14 and 0.002% to porcine pro-somatostatin (1–32), and itoes not recognize substance P, neuropeptide Y, human, bovine,orcine and rat vasoactive intestinal peptides, human insulin,uman amylin, and human, bovine, and porcine glucagon (1–29).he specificity of the anti-CRH and anti-somatostatin antibodiesas also supported by the expected well-known distribution of

abeled cell bodies in the hypothalamic paraventricular anderiventricular nuclei, respectively. Following incubation in pri-ary antibodies, the sections were washed, and CRH or soma-

ostatin was visualized by incubating the sections in Alexa Fluor94 anti-rabbit secondary antibody (Molecular Probes, 1:500) forh. Finally, sections were mounted and coverslipped, as de-

cribed above for single PTH2R labeling.

acaque tissue

issue collection and all procedures were performed according torotocols approved by the Animal Care and Use Committee of theational Institute of Mental Health following ethical review, and inccordance with the Institute for Laboratory Animal Research Guide

or the Care and Use of Laboratory Animals and conformed tonternational guidelines on the ethical use of animals. The authorsurther attest that all efforts were made to minimize the number ofnimals used and their suffering. A 3-day-old male macaqueonkey (Macaca mulatta) was sedated with ketamine and then

he testis was dissected for the preparation of hybridization probesnd the brainstem was dissected for in situ hybridization histochem-

stry. Tissues were frozen and stored at �80 °C until sectioning.Immunohistochemistry was performed on sections from a

-year-old male macaque that had been used for unrelated phys-ology experiments. Following deep pentobarbital anesthesia thisnimal was perfused transcardially with saline followed by 4%araformaldehyde in 0.1 M PB, at pH 7.4 (PB). Once the brain wasemoved, it was cryo-protected by first placing it into 10% glyce-ol�2% dimethyl-sulfoxide in PB for 24 h and then transferring itnto 20% glycerol�2% dimethyl-sulfoxide in PB for 5 days. Therain was then cut into blocks, frozen, and stored at �80 °C untilectioning.

Different aged monkeys were used for the in situ hybridizationnd immunolabeling experiment because we wanted to minimize theumber of animals euthanized. Since, in rats, the expression level ofIP39 is the highest during early postnatal period and no variationith age in PTH2R expression was observed (Dobolyi et al., 2006b),nd our experience with in situ hybridization is with fresh frozenaterial, a brain from a 3-day-old monkey was directly frozen for theetection of TIP39 and PTH2R mRNA by in situ hybridization. Webtained perfusion fixed material from the brain of a 9-year-old ma-aque already euthanized for unrelated purposes to avoid euthaniz-ng an animal for PTH2R immunolabeling.

acaque probe preparation for in situ hybridization

emplate cDNA was prepared from macaque testis by an RTeaction, as described above. PCR reactions were also performedssentially as described above using primers with human TIP395=-GGGGACTGTGCGGGAAGC-3= and 5=-GCATGTACGAGT-CAGCCAGTGG-3=) and PTH2R (5=-TGTGGGGCTTCATCTT-ATAGG-3= and 5=-ATGGCGGTGTCCTTTTCCAGTC-3=) se-uences. The PCR products were cloned into plasmid vectors,nd their identities were confirmed by DNA sequencing. A plasmidor each probe was used as template in PCR reactions, usingrimers that appended a T7 RNA polymerase recognition se-uence (5=-GCGCGTAATACGACTCACTATAGGG-3=) into the 5=nd of the antisense primer. The TIP39 probe was a 372 basemplicon, which differed from the predicted human sequence (Gen-ank accession number NM_178449) at 16 scattered base posi-

ions. It also lacked codons for two amino acid residues in theredicted leader sequence. The PTH2R probe was a 500 basemplicon. It differed from the predicted human sequence (GenBankccession number NM_005048) at 14 bases. It also contained a 49ase insertion that corresponds to predicted intronic sequence. Themall number of differences between the macaque probes and pre-icted human sequences may reflect PCR error or artifacts that doot affect their use as probes, or genuine species differences. Se-uencing was performed by the National Institute of Neurologicalisease Intramural DNA Sequencing Core Facility.

n situ hybridization histochemistry in macaque

erial coronal sections (12 �m) were cut from a macaque tissuelock using a cryostat. Sections were collected from the rostralorder of the diencephalon (about 1 mm rostral to the crossover

evel of the anterior commissure) to the caudal border of the ponsabout 17 mm caudal to the anterior commissure) mounted onositively charged slides (SuperfrostPlus®, Fisher Scientific, Pitts-urgh, PA, USA), dried, and stored at �80 °C until use. Theistance between sections used for mapping TIP39 and PTH2RRNA was 240 �m. Antisense (35S)UTP-labeled riboprobes wereenerated using T7 RNA polymerase from a MAXIscript transcrip-ion kit (Ambion, Austin, TX, USA) with TIP39 or PTH2R templatesescribed above. We have shown previously in rat that an anti-ense probe corresponding to the same region of TIP39 producesquivalent hybridization patterns to probes with non-overlapping

3apnripitsKTdGd

I

Spa0bawdimcbpfat(bHt2bgondcitp

M

SasocOswdUflfiE(owAt

ipa

Mh

HWoSRssuiwitdtbdtgtnn1ftrmnbctsg4e

Dh

Pb

a

Pbnsgc

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147132

7 (Dobolyi et al., 2003b). Similarly, we have shown in rat that thentisense probe corresponding to the same region of PTH2Rroduces equivalent hybridization patterns to probes with theon-overlapping sequence corresponding to bases 1274–1828 inat (Wang et al., 2000). For hybridization, we used 80 �l hybrid-zation buffer and 1 million disintegration per minute labeled probeer slide. The in situ hybridization protocols have been described

n detail, elsewhere (http://intramural.nimh.nih.gov/lcmr/snge/Pro-ocols/ISHH/ISHH.html). Following hybridization and washes,lides were dipped in NTB nuclear track emulsion (Eastmanodak, Rochester, NY, USA) and stored at 4 °C for 3 weeks.hen, the slides were developed and fixed with Kodak Dektoleveloper and Kodak fixer, respectively, and counterstained withiemsa stain, and coverslipped with Cytoseal 60 mounting me-ium (Stephens Scientific, Riverdale, NJ, USA).

mmunolabeling in macaque tissue

ections from the 9-year-old macaque had been cut in the coronallane on a sliding microtome at a thickness of 40 �m and storedt �20 °C in a solution of 30% ethylene glycol�20% glycerol in.05 M PB. For double labeling PTH2R and VGLUT2 immunola-eling was performed on sections extending from the level of thenterior commissure to the end of the hypothalamus, essentiallyith the same procedure as for single labeling of the PTH2Rescribed above, followed by antigen retrieval and fluorescent

mmunolabeling of the VGLUT2. Briefly, after FITC-tyramide im-unolabeling of the PTH2R, the sections were placed in 10 mM

itric acid buffer, pH 6.0 for 30 min at room temperature followedy incubation at 90 °C for 15 min. Subsequently, the sections werelaced in anti-VGLUT2 primary antiserum (1:10,000 dilution; a gift

rom Dr. Robert Edwards, UCSF, CA, USA) for 48 h, at 4 °C. Thisntiserum was developed against a fusion protein containing glu-

athione S-transferase and the carboxy-terminal 64 amino acidsresidues 519–582) of rat VGLUT2 (Fremeau et al., 2001). It haseen used and characterized extensively (Fremeau et al., 2001;artig et al., 2003) and Western blot analysis demonstrated that

his antibody specifically recognizes VGLUT2 (Fremeau et al.,001). Following the incubation in the anti-VGLUT2 primary anti-ody, the sections were incubated in carbocyanine (Cy)-5–conju-ated anti-rabbit secondary antibody (1:300 dilution; Vector Lab-ratories) for 5 h followed by incubation in 4=,6-diamidino-2-phe-ylindole (DAPI) for 5 min. The sections were then mounted,ried, and coverslipped in polyvinyl alcohol (PVA)–1,4-diazabicy-lo-2,2,2-octane (DABCO) antifade mounting medium. Control

mmunostaining, in which the VGLUT2 primary antibody was omit-ed, was used to verify that the Cy-5 labeling derives from theresence of VGLUT2.

icroscopic analysis

ections labeled by in situ hybridization histochemistry were ex-mined using a Zeiss Axioplan 2 and an Olympus IX70 micro-cope. Photomicrographs were captured at 1300�1030 pixel res-lution with an AxioCam HR camera or a Roper CoolSnap FXamera. Montages were created using Zeiss AxioVision software.ther sections were examined using an Olympus BX60 micro-cope also equipped with fluorescent epi-illumination. Imagesere captured at 2048�2048 pixel resolution with a SPOT Xplorerigital CCD camera (Diagnostic Instruments, Sterling Heights, MI,SA). Photomicrographs were taken using a 4� objective foruorescent and dark-field, and 20� or 40� objectives for bright-eld images. Confocal images were acquired with a Nikon Eclipse800 confocal microscope equipped with a Bio-Rad Radiance

2100) Laser Scanninig System using a 60� objective at anptical thickness of 2 �m. Contrast and sharpness of the imagesere adjusted using the “levels” and “sharpness” commands indobe Photoshop CS 8.0. The full resolution was maintained until

he micrographs were cropped and assembled, at which point t

mages were adjusted to a resolution of 300 dpi. Drawings wererepared by aligning the pictures with corresponding schematicsdapted from a macaque brain atlas (Martin and Bowden, 1996).

RESULTS

apping of the expression of the PTH2R in theuman brain by RT-PCR

uman brain tissue samples yielded 2–10 �g total RNA.hen run on denaturing formaldehyde gels, the intensity

f the 28 S rRNA band was at least equal to that of the 18rRNA band in all RNA samples, suggesting acceptable

NA quality. When these RNA samples were reverse tran-cribed and amplified using PTH2R specific primer pairs, aingle band appeared, as shown in Fig. 1. The PCR prod-ct length corresponded to the expected 440 bp suggest-

ng appropriate specific PCR products for the primer pair,hich was also confirmed by sequencing. A band with high

ntensity signal was found for samples from the septum,he caudate nucleus, the medial geniculate body, the me-ial hypothalamus, the pretectal area, the pontine tegmen-um, and the cerebellar cortex (Fig. 1). A lower intensityand was found for the frontal cortex, hippocampus, amyg-ala, lateral geniculate body, subthalamic nucleus, ventralegmental area, dorsal vagal complex, and the spinal tri-eminal nucleus (Fig. 1). In contrast, no band was de-ected for the ventral thalamus, the mediodorsal thalamicucleus, the pulvinar, the substantia nigra, the pontineuclei, the ventrolateral medulla, and the inferior olive (Fig.). There were nine brain regions analyzed in samplesrom two different brains. Samples from the hippocampus,he subthalamic nucleus, and the pontine reticular formationesulted in indistinguishable PCR bands. The ventral thala-us, the mediodorsal thalamic nucleus, and the pontineuclei did not contain detectable PTH2R mRNA in eitherrain. The pretectal area, the pontine tegmentum, and theerebellar cortex were also all positive for PTH2R buthe intensity of the band was lower in the samples from theecond brain. Using primers specific for the housekeepingene GAPDH resulted in bands of the expected length of23 bp with about the same intensity for all the samplesxcept for the no RNA negative control (Fig. 1).

istribution of PTH2R immunoreactivity in theuman brain

TH2R-ir fibers had a widespread distribution in severalrain areas, as described below (Table 2).

Cerebral cortex. The insular cortex was examinednd only a few scattered PTH2R-ir fibers were detected.

Thalamus. The thalamus had a relative paucity ofTH2R-ir fibers (Table 2). High densities of PTH2R-ir fi-ers were only detected in the paraventricular thalamicucleus (Fig. 2H). In addition, a few PTH2R-ir fibers werecattered in the anteromedial, lateral, posterior, and supra-eniculate thalamic nuclei, in the habenula, the peripedun-ular area, and the lateral geniculate body.

Hypothalamus. In general, the hypothalamus con-

(wtlati(tstdtawi

mfwdbPnahPt

l

ssfipt

finfidudwr

Iitldp

Ra

Slth

Fhs ds of 42t

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 133

Table 2). A particularly dense network of PTH2R-ir fibersas seen in the medial preoptic area (Fig. 2B), in contrast to

he low density in the lateral preoptic area. The paraventricu-ar (Fig. 2D) and infundibular nuclei (Fig. 2E) also contained

very dense fiber network, with PTH2R-ir fibers projectingowards the median eminence where very intense PTH2Rmmunolabeling was visible, particularly in the external layerFig. 2F). Immunopositive fibers were also numeroushroughout the entire lateral hypothalamic area (Fig. 2I), theupraoptic (Fig. 2B), dorsomedial, posterior hypothalamic,uberomamillary and premamillary nuclei. A moderate to highensity of PTH2R-ir fibers was present in the anterior hypo-

halamic, ventromedial, perifornical, and subthalamic nuclei,s well as in the nuclei of the mamillary body (Fig. 2I),hereas only a few scattered PTH2R-ir fibers were detected

n the suprachiasmatic nucleus (Fig. 2B).

Midbrain. The highest density of PTH2R-ir fibers in theidbrain was within the periaqueductal gray. PTH2R-ir fibers

ormed networks in the dorsomedial and lateral subdivision,hereas the number of PTH2R-ir fibers was lower in theorsolateral and ventral subdivisions (Fig. 3D). Other mid-rain regions that contained a relatively high density ofTH2R-ir fibers include the zona incerta, the subbrachialucleus, the pretectal area, the superior and inferior colliculi,nd the ventral tegmental area. The dorsal raphe nucleusad a considerably lower density of PTH2R-ir fibers, but noTH2R immunoreactivity was present in the substantia nigra,

he red nucleus, and the oculomotor nuclei.

Pons. PTH2R-ir fibers were densely packed in the

ig. 1. Expression of the PTH2R in the human brain detected by RT-Puman brain regions were run on gels. In the upper line, PTH2R-specuggest a widespread expression of the PTH2R. In the lower line, banhe housekeeping gene GAPDH are present.

ateral parabrachial nucleus (Fig. 3B). A much lower den- i

ity of PTH2R-ir fibers was seen in the tegmentum and theensory trigeminal nuclei. In addition, solitary PTH2R-irbers were scattered in the pontine reticular formation, theontine raphe nucleus, the nuclei of the superior olive andhe vestibular nuclei.

Medulla oblongata. A very dense network of PTH2R-irbers was seen in the marginal layer of the spinal trigeminalucleus (not shown). In addition, a high density of PTH2R-irbers was present in the nucleus of the solitary tract imme-iately adjacent to the area postrema and in the dorsal retic-lar nucleus of the medulla. Some PTH2R-ir fibers were alsoistributed in the area postrema. In addition, scattered fibersere present in the different compartments of the medullary

eticular formation.

Spinal cord. In sections of the spinal cord, laminaeand II in the dorsal horn showed strikingly intense

mmunolabeled varicose fibers (Fig. 3E). Lamina X andhe intermediomedial nucleus contained a few lightlyabeled fibers. Scattered fibers were present in all of theeep layers of the dorsal horn and a few fibers were alsoresent in the ventral horn.

elationship between the PTH2R and somatostatin-,nd CRH-containing hypothalamic neurons in human

omatostatin-ir perikarya were distributed in the hypotha-amic periventricular nucleus surrounded by somatosta-in-ir fibers projecting ventrally towards the medial basalypothalamus and the median eminence (Fig. 4A). Double

ucts of PCR reactions performed using cDNA templates from differentp PCR products are visible. The positive bands in several brain areas3 bp from control PCR reactions performed using primers specific for

CR. Prodific 440 b

mmunolabeling revealed that a portion of these somatosta-

te

P4ehtfii

hifiHp

Db

oea

PphaceaamrteP

mmPmlsdp

aAaAAaBCccDDdDDDDDDEfIILLlLLLmMmMmMMMM

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147134

in-ir fibers contain PTH2R-ir while other fibers containedither only somatostatin, or only PTH2R-ir (Fig. 4A, B).

A very high density of both somatostatin-ir andTH2R-ir fibers was present in the median eminence (Fig.C). The majority of PTH2R-ir fiber terminals in the medianminence contained somatostatin, as demonstrated byigh magnification images of double-immunostained sec-ions (Fig. 4C). In contrast, only a portion of somatostatin-irber terminals in the median eminence showed co-local-zation with PTH2R-ir (Fig. 4C).

CRH-ir cell bodies and fibers were distributed in theypothalamic paraventricular nucleus (Fig. 4D, E). Double

mmunolabeling demonstrated that CRH-ir cell bodies andbers do not contain PTH2R immunoreactivity (Fig. 4E).owever, PTH2R-ir fiber terminals closely apposed CRH-irerikarya (Fig. 4E).

istribution of PTH2R mRNA in the macaque brainy in situ hybridization histochemistry

Basal ganglia and other forebrain structures. Somef the most rostral sections allowed us to analyze PTH2R-xpressing neurons in the hippocampus, the amygdala,nd some of the basal ganglia. Only a few scattered

Abbreviation

c anterior commissurecN accumbens nucleusl ansa lenticularisM anteromedial thalamic nucleusPr anteroprincipal thalamic nucleusq cerebral aqueductNST bed nucleus of the stria terminaliseA central amygdaloid nucleusp cerebral peduncleu cuneate fasciculusG dentate gyrusH dorsal hornlf dorsolateral fasciculusLPAG periaqueductal gray, dorsolateral subdivisionMH hypothalamic dorsomedial nucleusMPAG periaqueductal gray, dorsomedial subdivisionpMe deep mesencephalic nucleusR dorsal raphe nucleusTg dorsal tegmental nucleusGP external globus pallidus

r fasciculus retroflexusGP internal globus pallidusnf infundibular nucleusG lateral geniculate bodyHA lateral hypothalamic area

l lateral lemniscusPAG periaqueductal gray, lateral subdivisionV lateral ventricle2 lamina 2 of the spinal dorsal horne median eminenceeA medial amygdaloid nucleusfb medial forebrain bundleG medial geniculate bodylf medial longitudinal fasciculusMG medial nucleus of the medial geniculate bodyMN medial mamillary nucleusPA medial preoptic area

TH2R-expressing neurons were visible in the hippocam-us, distributed throughout the dentate gyrus and theippocampal areas. In contrast, the amygdala containedvery high density of PTH2R-expressing neurons in the

entral amygdaloid nucleus (Fig. 5C), while PTH2R-xpressing neurons were also abundant in the medialmygdaloid nucleus (Fig. 5C). Other regions of themygdala contained significantly lower levels of PTH2RRNA. Also, a low density of PTH2R-expressing neu-

ons was found in the caudate nucleus and the substan-ia innominata while the putamen and the internal andxternal parts of the globus pallidus did not containTH2R mRNA (Fig. 5C).

Thalamus. In general, the thalamus showed veryodest PTH2R mRNA labeling with the exception of itsetathalamic portion, the medial geniculate body. Here,TH2R-expressing neurons had a very high density in theedial nucleus of the medial geniculate body, a somewhat

ower density in the ventral nucleus (Fig. 5E), and adjacentuprageniculate thalamic nucleus and no labeling in theorsal nucleus. In addition, PTH2R-expressing cells wereresent, at low density, in the midline thalamic nuclei,

the figures

N medial preoptic nucleusmotor oculomotor nucleusoptic tracthypothalamic paraventricular nucleus

G periaqueductal grayG parabigeminal nucleusL lateral parabrachial nucleus

hypothalamic periventricular nucleuspiriform cortexpontine nuclei

O pontine reticular nucleusperipeduncular nucleusputamen

l pulvinar thalamiT paraventricular thalamic nucleus

pyramidal tractreticular thalamic nucleus

g sagulum nucleussuperior colliculus

h suprachiasmatic nucleussuperior cerebellar pedunclesubstantia innominatastria medullarissubstantia nigrasupraoptic nucleus

C superior olivary complexF subparafascicular areaF subparafascicular nucleusFp subparafascicular nucleus, parvicellular part

ventroanterior thalamic nucleusL ventral nucleus of the lateral lemniscusG ventral nucleus of the medial geniculate bodyH hypothalamic ventromedial nucleusL ventral posterolateral thalamic nucleus

third ventricletrochlear nerve

s used in

MPnuOtPaPAPBPBpePirPnPnPPPuPuPVpyRtSaSCSCscpSIsmSNsoSOSPSPSPVAVLVMVMVP3V4n

T

A

F

B

D

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 135

able 2. Summary of the location of PTH2R mRNA and PTH2R-ir fibers in human and macaque, and their comparisons to mouse data

reamRNA by RT-PCRin human

mRNA by ISHHin macaque

Cell bodiesin mice

PTH2R-ir fibersin human

PTH2R-ir fibersin mice

orebrainCerebral cortex

Frontal cortex � � 0Insular cortex �� � �

Hippocampus � � � 0Septum ���

Medial septal nucleus 0 0Lateral septal nucleus ��� ��

Amygdala �

Central nucleus ��� �� ��

Basal nuclei � � 0Lateral nucleus � � 0Medial nucleus �� �� ��

Cortical nucleus � � �

asal gangliaCaudate nucleus ��� � �� 0Putamen 0Globus pallidus 0 � 0Claustrum �� �

Nucleus accumbens � ��

Substantia innominata � � �

iencephalonThalamus

Anterior thalamic nuclei 0 0 � 0Midline thalamic nuclei � �� �� ��

Lateral thalamic nuclei 0 0 � �

Ventral thalamic nuclei 0 0 0 0 0Reticular nucleus 0 0 0 0Mediodorsal thalamic nucleus 0 0 0 0 0Habenular nuclei � � � �

Posterior thalamic nuclei 0 0 � �

Peripeduncular area � � � �

Suprageniculate thalamic nucleus � � � �

Medial geniculate body ���

Dorsal nucleus 0 0 0 0Ventral nucleus �� �� � �

Medial nucleus ��� ��� �� ��

Lateral geniculate body � � � � �

Pulvinar 0 0Hypothalamus ���

Medial preoptic area ��� ��� ��� ���

Lateral preoptic area � � � �

Supraoptic nucleus � � �� ��

Supraoptic decussations �� ��

Suprachiasmatic nucleus 0 0 0 0Anterior hypothalamic nucleus � � � �

Paraventricular nucleus ��� �� ��� ���

Periventricular nucleus �� �� �� ��

Arcuate (infundibular) nucleus � �� ��� ���

Median eminence ��� ���

Ventromedial nucleus � � � �

Dorsomedial nucleus � � �� ��

Lateral hypothalamic area �� � �� ��

Perifornical nucleus � � � �

Posterior hypothalamic nucleus � � �� ��

Tuberomamillary nucleus � � �� ��

Premamillary nuclei � � �� ��

pltam

hvi

T

A

M

SB

P

M

C

Pis

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147136

articularly in the paraventricular thalamic nucleus, theateral habenular nucleus, the peripeduncular area, andhe lateral geniculate body. The anterior, lateral, ventral,nd posterior thalamic cell groups did not show PTH2R

able 2. continued

reamRNA by RT-PCRin human

amillary bodySuperior mamillary nucleusMedial mamillary nucleusLateral mamillary nucleus

ubthalamic nucleus ��

rainstemMidbrain

Zona incertaSubstantia nigra 0Red nucleusSubbrachial nucleusPraetectal area ���

Superior colliculusInferior colliculusParabigeminal nucleusPeriaqueductal grayDorsal raphe nucleusVentral tegmental area �

Oculomotor nucleionsLateral lemniscal nucleiMedial paralemniscal nucleusPontine tegmentum. ���

Parabrachial nucleiMedialLateralPontine nuclei 0Superior olivePontine reticular formation �

Principal trigeminal nu.Motor trigeminal nucleusPontine raphe nucleusVestibular nucleiedulla oblongataCochlear nucleiSpinal trigeminal nucleus �

Prepositus hypoglossal nucleusMedullary reticular formation 0Inferior olive 0Dorsal vagal complex �

Nucleus of the solitary tractDorsal motor vagal nucleusArea postremaMotor hypoglossal nucleusNucleus ambiguusMedullary raphe nucleierebellumCortex ���

Nuclei

The average number of labeled cells per section in the nucleus isTH2R-ir fibers and fiber terminals is represented as none to low (

mmunolabeling in the hypothalamus and medulla oblongata is determtudy (Faber et al., 2007).

RNA labeling (Table 2). p

Hypothalamus. In situ hybridization indicated that theypothalamus was rich in PTH2R-expressing neurons. Aery high density of PTH2R-expressing neurons was found

n the medial preoptic nucleus while other regions of the

y ISHHue

Cell bodiesin mice

PTH2R-ir fibersin human

PTH2R-ir fibersin mice

�� � ��

0 0 00 � ��

� � �

�� �� ���

0 0 00 0 0� �� ��

� �� ��

�� �� ��

� �� ��

�� ��� ���

�� �� ���

�� �� ���

0 0 0

0 0 00 � �

�� �� ��

�

� �

��� ���

0 0 00 � �

0 � �

� �� ��

0 0 00 � ��

0 � �

0 0� ��� ���

0 0 00 � �

0 0 0

�� �� ��

0 0 00 0 00 0 00 0 00 0 0

�� �

0 0

1–10 (�), 11–20 (��), or over 20 (���). Similarly, the density ofrate (�), high (��), and very high (���). The density of PTH2Rthe average of two brains. The mice data are based on our previous

mRNA bin macaq

�

00�

�

00�

�

���

�

���

��

���

�

0

00��

�

00�

�

000

none (0),0), modeined as

reoptic area exhibited a low density of labeled neurons

Ftdt

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 137

ig. 2. Immunolabeling of the PTH2R in the human diencephalon. Photomicrographs demonstrate PTH2R immunoreactivity visualized by FITC-yramide fluorescent amplification immunocytochemistry. The presence of PTH2R-ir fibers is demonstrated in coronal sections of the human

he field in B. (B) A dense network of PTH2R-ir fibers is present in the medial preoptic area. In contrast, the density of PTH2R-ir fibers is very low in

(in5p

nette

ttPAbaaab

Fbctmigtd em (Paxa as zero.

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147138

not shown). A very high intensity of PTH2R mRNA label-ng was also seen in the hypothalamic paraventricularucleus, particularly in its parvicellular subdivisions (Fig.B). Labeled neurons extended into the hypothalamiceriventricular nucleus, albeit in a significantly smaller

he supraoptic nucleus, and PTH2R-ir fibers are absent in the suprachhe level of the paraventricular nucleus in the coronal plane. The framTH2R-ir fibers is present in the hypothalamic paraventricular nucleuvery high level of PTH2R immunoreactivity is present in fiber termina

rain at the level of the mamillary body in the coronal plane. (H) PTH2djacent anteromedial thalamic nucleus. (I) The density of PTH2R-ir fibsent from the medial mamillary nucleus. Schematic drawings (A, C

ig. 3. Immunolabeling of the PTH2R in the human brainstem and spy FITC-tyramide fluorescent amplification immunocytochemistry in cooronal plane. The framed area indicates the field in B. (B) A high dehe adjacent sagulum nucleus and the ventral nucleus of the lateralesencephalic section of the human brain in coronal plane. The framed

n the periaqueductal gray. A particularly high density of PTH2R-ir fiberay. (E) A particularly dense network of PTH2R-ir fibers is present inhe cuneate and dorsolateral fasciculi, and the deeper laminae of therawings (A, C) are modifications from an atlas of the human brainstgreement with the scaling of this atlas, with the level of the obex set

greement with the scaling of this atlas, anteroposterior coordinates (AP) are inars�1 mm for B, H, and I, and 500 �m for D, E, and F.

umber (Fig. 5B). A cluster of intensely labeled PTH2R-xpressing neurons was also detected in the ventral part ofhe lateral hypothalamic area (Fig. 5B). Many other parts ofhe hypothalamus showed only a low density of PTH2R-xpressing neurons including the supraoptic, anterior hy-

nucleus. (C) A schematic drawing shows a part of the human brain ats correspond to fields in D and E, respectively. (D) A high density ofe infundibular nucleus contains a high density of PTH2R-ir fibers. (F)edian eminence. (G) A schematic drawing shows a part of the human

rs are abundant in the paraventricular thalamic nucleus but not in theigh in the lateral hypothalamic area. In contrast, PTH2R-ir fibers aremodifications from an atlas of the human brain (Mai et al., 1997). In

. Photomicrographs demonstrate PTH2R immunoreactivity visualizedtions (A) A schematic drawing shows a section of the human pons inTH2R-ir fibers is present in the lateral parabrachial nucleus whereass do not contain PTH2R-ir fibers. (C) A schematic drawing shows aicates the field in D. (D) A dense network of PTH2R-ir fibers is presentent in the dorsomedial and lateral subdivisions of the periaqueductal

of the dorsal horn as demonstrated at cervical segment 2. In contrast,rn do not contain more than few, sparse PTH2R-ir fibers. Schematic

inos and Huang, 1995). Anteroposterior coordinates are indicated inScale bars�500 �m for B, D, and E.

iasmaticed area

s. (E) Thls of the mR-ir fibebers is h, G) are

inal cordronal secnsity of Plemniscu

area indrs is pres

lamina 2dorsal ho

dicated from the level of the anterior commissure set as zero. Scale

pfmtnm

scil

FaSbpoefiPnat

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 139

othalamic, infundibular, ventromedial, dorsomedial, peri-ornical, posterior hypothalamic, tuberomamillary, and pre-amillary nuclei. PTH2R-expressing cells were scarce in

he mamillary body and the subthalamic nucleus. Someeurons of the superior but none in the medial or lateral

ig. 4. PTH2R-ir fibers in relation to somatostatin and CRH immunorend in the median eminence. Photomicrographs demonstrate PTH2R-omatostatin-ir cell bodies and fibers (red), PTH2R-ir fibers (green) indy white arrowheads are present in the hypothalamic periventricularortion of the somatostatin neurons contain PTH2Rs. (B) The framed anly somatostatin immunoreactivity (red) as well as fibers in which somminence in a high magnification confocal image. White arrowheads ibers are located in the hypothalamic paraventricular nucleus. The fraTH2R-ir fibers and fiber terminals (green) are shown in a high magnio co-localization between CRH and the PTH2R. In turn, PTH2R-ir fibrrows. Scale bars�200 �m for A, 50 �m for B and E, 30 �m for C, anhe reader is referred to the Web version of this article.

amillary nuclei contained PTH2R mRNA. w

Midbrain. PTH2R-expressing cells were abundant ineveral different midbrain regions. The superior colliculusontained a very high density of PTH2R-expressing cells in

ts superficial layers and fewer labeled neurons in the deepayers (Fig. 6B). Similarly, the PTH2R-expressing neurons

in the human hypothalamic periventricular and paraventricular nucleicoronal sections labeled for somatostatin (A–C) and CRH (D, E). (A)black arrowheads, as well as double-labeled fibers (yellow) indicatedThe co-localization of somatostatin and the PTH2R suggests that a

is shown in a high magnification confocal image. (C) Fibers containingand the PTH2R co-localize (yellow), are demonstrated in the median

ome of the double-labeled fiber terminals. (D) CRH-ir cell bodies anda indicates the field in E. (E) CRH-ir cell bodies and fibers (red) andonfocal image. The lack of double-labeled yellow structures suggestsals seem to closely appose CRH-ir cell bodies as indicated by whiteor D. For interpretation of the references to color in this figure legend,

activitiesir fibers inicated bynucleus.rea in Aatostatin

ndicate smed are

fication cer termind 1 mm f

ere topographically organized in the periaqueductal gray

asfrspeeeznwo

erwtf

Vt

Iec

Fipmsadthagaa den, 199w set as z

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147140

s well. The lateral subdivision demonstrated a high den-ity of labeled neurons while other subdivisions containedewer PTH2R-expressing neurons (Fig. 6C). The dorsalaphe nucleus also had a strong PTH2 receptor mRNA initu hybridization signal (Fig. 6D). In addition, denselyacked PTH2R-expressing neurons were apparent in thentire length of the parabigeminal nucleus (Fig. 6E). Sev-ral other regions of the midbrain contained a low to mod-rate density of PTH2R-expressing cells, including theona incerta, the ventral tegmental area, the subbrachialucleus, and the inferior colliculus. No labeled neuronsere found in the substantia nigra, the red nucleus, or the

ig. 5. Expression of the PTH2R in diencephalic and amygdaloid regin situ hybridization histochemistry in coronal sections of the macaqueresence of TIP39 mRNA in the dark-field photomicrographs of emulacaque brain through the hypothalamus and the amygdala. The fra

hows a high density of PTH2R-expressing neurons in the hypothalarea. (C) The dark-field photomicrograph shows a very high densityensity of PTH2R-expressing neurons in the medial amygdaloid nucle

he level of the medial and lateral geniculate bodies. The framed areaigh density of PTH2R-expressing neurons in the medial nucleus obundant in the ventral nucleus of the medial geniculate body. In coeniculate body. The arrow points to the field in F. (F) A high magnificutoradiography grains above cell bodies containing PTH2R mRNA inre modifications from an atlas of the macaque brain (Martin and Bowith the scaling of this atlas, with the level of the anterior commissure

culomotor nuclei (Table 2). a

Pons. The pons had few PTH2R-expressing neuronsxcept for the dorsal tegmental nucleus, which contained aelatively high level of PTH2R mRNA. Labeled neuronsere sparsely distributed in other parts of the pontine

egmentum, the parabrachial nuclei, the pontine reticularormation, and the sensory trigeminal nucleus.

GLUT2 immunoreactivity in PTH2R-containing fibererminals in macaque

n the regions of the macaque brain examined, whichxtended from the level of the anterior commissure to theaudal end of the diencephalon, PTH2R-ir fibers were

macaque. The presence of mRNA encoding the PTH2R is shown bysitive autoradiography signals appear as white dots demonstrating theed material. (A) A schematic drawing shows a coronal section of thes indicate the fields in B and C. (B) The dark-field photomicrographentricular nucleus and in the ventral part of the lateral hypothalamic-expressing neurons in the central amygdaloid nucleus and a lowerschematic drawing shows a coronal section of the macaque brain at

nds to the field in E. (E) The dark-field photomicrograph shows a verydial geniculate body while the PTH2R-expressing neurons are alsonly few scattered PTH2R expressing cells are present in the lateralht-field photomicrograph demonstrates the accumulation of individualial geniculate body. Schematic drawings in all macaque brain figures6), and anteroposterior coordinates (AP) are indicated in agreementero. Scale bars�1 mm for B, C, and E, and 30 �m for F.

ons of thebrain. Posion dippmed areamic paravof PTH2Rus. (D) Acorrespof the mentrast, o

ation brigthe med

bundant in the ventral part of the lateral septal nucleus

amfi

idt

FhPiiPPpp diographye and 100

FdnidasPi

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 141

nd in the same regions of the thalamus and hypothala-us, as described in human. In addition, some scattered

bers were present in cortical areas. In some regions,

ig. 6. Expression of the PTH2R in the macaque midbrain. Theistochemistry in coronal sections of the macaque midbrain. PositiveTH2R mRNA in the dark-field photomicrographs. (A) A schematic d

ndicate the fields in the dark-field photomicrographs in B–E. (B) The dan the superior colliculus. The density of labeled neurons is especiaTH2R-expressing neurons in the periaqueductal gray. The densityTH2R-expressing neurons are abundant in the dorsal raphe nucarabigeminal nucleus lateral to the lateral lemniscus. The framedhotomicrograph demonstrates the accumulation of individual autoraminal nucleus. Scale bars�1 mm for B and C, 500 �m for D and E,

ig. 7. PTH2R-ir fibers, in relation to VGLUT2-immunoreactivity in aemonstrates that PTH2R-ir (green) fibers as well as VGLUT2-ir (reucleus, the medial preoptic nucleus, and the adjacent preoptic regio

ndicates the position of the field in B and C. (B) The co-localizationemonstrates that the majority of PTH2R-ir fiber terminals also co-locarrowheads. In contrast, there are numerous VGluT2-containing fiberame field as in B is shown without the green channel to demonstrate

n this figure legend, the reader is referred to the Web version of this article.

ncluding the lateral septum and preoptic area (Fig. 7)ouble-labeling revealed that essentially all PTH2R-ir fibererminals contained VGLUT2-ir.

of PTH2R encoding mRNA is exhibited by in situ hybridizationgraphy signals appear as white dots to demonstrate the presence ofhows a coronal section of the macaque midbrain. The framed areashotomicrograph shows that PTH2R-expressing neurons are abundantin the superficial layers. (C) The dark-field photomicrograph showsled neurons is especially high in the ventrolateral subdivision. (D)) A high density of PTH2R-expressing neurons is present in theorresponds to the field in F. (F) A high magnification bright-field

grains above cell bodies containing PTH2R mRNA in the parabig-�m for F.

ection of the macaque medial preoptic area. (A) A photomicrographand fiber terminals are present in the hypothalamic periventricularuclei of cell bodies are labeled blue with DAPI counterstaining. Star

R- and VGLUT2-ir (yellow) in the high magnification confocal imageUT2. Several of these double-labeled terminals are indicated by whites that do not contain PTH2R as indicated by their red color. (C) Thence of VGLUT2 immunoreactivity (red) in fiber terminals co-localizing

presenceautoradiorawing srk-field plly high

of labeleus. (E

area c

coronal sd) fibersns. The nof PTH2lize VGLterminal

the prese

Di

Ntlcrwtanab

clTsi2

TiTbnciarTtsmara

FcdTtffnb ual autorb

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147142

istribution of TIP39 mRNA in the macaque brain byn situ hybridization histochemistry

umerous TIP39 mRNA containing neurons were found inhe posterior thalamus. One group of these neurons wasocated in the subparafascicular area between the mostaudal portion of the third ventricle and the fasciculusetroflexus (Fig. 8B). Additional TIP39-expressing neuronsere found lateral to the fasciculus retroflexus (Fig. 8B) in

he parvicellular subparafascicular nucleus distributinglong the dorsal edge of the medial lemniscus. Theseeurons extended laterally as far as the peripeduncularrea immediately ventromedial to the medial geniculateody.

Another group of TIP39-expressing neurons was lo-ated medial to the dorsal part of the ventral nucleus of theateral lemniscus at the midbrain-pons junction (Fig. 8D).he topography of these TIP39-expressing neurons corre-ponds to the medial paralemniscal nucleus, as has beendentified in rodents (Dobolyi et al., 2003b; Varga et al.,

ig. 8. Expression of TIP39 in the macaque brain. The presence oforonal sections of the macaque brain. Positive autoradiography signark-field photomicrographs. (A) A schematic drawing shows a coronahe framed area indicates the field in B. (B) The dark-field photomicro

he white matter of the fasciculus retroflexus in the subparafasciculaasciculus retroflexus. (C) A schematic drawing shows a coronal sectiramed area corresponds to the field in D. (D) The dark-field photomicucleus immediately medial to the ventral nucleus of the lateral lemniright-field photomicrograph demonstrates the accumulation of individars�1 mm for B and D, and 200 �m for E.

008). m

DISCUSSION

he major purpose of this study was to provide a compar-son of the neuroanatomical distribution of the PTH2R andIP39 in human and non-human primates with their distri-utions in rodents. We are experimentally evaluating thiseuromodulator system in rodents and performed thisomparison to help evaluate the potential extrapolation of

ts functions in rodents to humans. We focused on materialnd brain regions most likely to be informative in thisegard. The major finding is that both the PTH2R andIP39 have a similar distribution in human and macaque to

heir distributions in rodents. Their expression is greatest inubcortical structures. The data support the idea that theyay be involved in similar functions. However, judgmentbout the relevance of particular observations made inodents to humans obviously requires detailed consider-tion of the relevant functions and structures.

In the following discussion the distribution of PTH2R

ncoding TIP39 is exhibited by in situ hybridization histochemistry inar as white dots to demonstrate the presence of TIP39 mRNA in theof the macaque brain at the diencephalon–mesencephalon junction.

ows TIP39-expressing neurons located between the 3rd ventricle andwell as in the parvicellular subparafascicular nucleus lateral to the

macaque brain at the level of the ponto-mesencephalic junction. Thehows TIP39-expressing neurons located in the medial paralemniscal

e framed area corresponds to the field in E. (E) A high magnificationadiography grains above cell bodies containing TIP39 mRNA. Scale

mRNA eals appel section

graph shr area, ason of therograph sscus. Th

RNA obtained with different methods and in different

mmnltpi

Dp

PRhnsdwms

ibtbFsiIGnsuwPteutnpctiwlrpivpsfivgbb

ts

PbomPidedttteivttscfiattPfhaftttgsthtn

pspentetmvatmbr

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 143

aterial is compared. Next, the distribution of PTH2R im-unoreactivity is compared between the human andon-human primate material used in this study and pub-

ished rodent data. Finally, the expression of TIP39 andhe presence of TIP39-PTH2R neuromodulator system inrimates are discussed with a view of potential functional

mplications.

istribution of the expression of PTH2R mRNA in therimate brain

TH2R mRNA was detected with two different methods,T-PCR in the brain of adults and in situ hybridizationistochemistry in 3-day old macaque monkey. The ratio-ale for comparing these distributions is that the expres-ion level of the PTH2R did not change during postnatalevelopment in the rat (Dobolyi et al., 2006b). However,e consider some methodological issues of detectingRNA of the PTH2R before comparing RT-PCR and in

itu hybridization data.

Methodological considerations. The RT-PCR exper-ments were designed so that the appearance of a PTH2Rand in the gel means that PTH2R mRNA was present inhe microdissected tissue sample while the absence of aand is a good indication of the absence PTH2R mRNA.irst, the mRNA quality was checked for each humanample to eliminate the possibility that mRNA degradationn the postmortem period would lead to a negative signal.n addition, the presence of the housekeeping geneAPDH was shown in each sample using a lower cycleumber to demonstrate that mRNA was efficiently tran-cribed into cDNA. In contrast, a high cycle number wassed for the PTH2R so that even a small amount of mRNAould be detected. For the better comparison of theTH2R expression in different samples, the RNA concen-

rations were measured and adjusted so 2 �g was used forach sample. A negative control sample without RNA wassed to exclude the possibility of contamination at eitherhe RNA or the cDNA level. Furthermore, an intron span-ing PTH2R primer pair was used to create larger PCRroduct amplified from genomic DNA so potential genomicontamination would be recognized. Finally, the specifici-ies of the PCR products were demonstrated by sequenc-ng. The reproducibility of our data were also confirmedhen nine samples of the same brain regions were ana-

yzed from two different human brains. Indistinguishableesults were obtained for six samples while three samplesroduced clear bands from both brains that differed in their

ntensity. Since PTH2R expression does not obviouslyary with gender or age in rat (Dobolyi et al., 2006b), it isossible that methodological reasons account for thesemall differences. For example, microdissection from dif-erent parts of the same nucleus could result in differentntensity PTH2R bands if the expression level of PTH2Raries topographically within a single brain nucleus or re-ion. Such topographic differences were indeed revealedy in situ hybridization in the macaque brain in various

he medial geniculate body, the periaqueductal gray, theuperior colliculus, and the pontine tegmentum.

Comparison of RT-PCR and in situ hybridization data.TH2R mRNA has a widespread expression in the brain ofoth human and non-human primates based on the resultsf both the RT-PCR and in situ hybridization experi-ents (Table 1). Cortical structures seem to contain someTH2R mRNA while the PTH2R expression level was high

n the septum and the caudate nucleus based on RT-PCRata in human. In the amygdala, RT-PCR showed a mod-rate level of PTH2R expression while in situ hybridizationemonstrated high levels of PTH2R expression in the cen-ral and medial nuclei, with only low levels in other nuclei ofhe amygdala. Thalamic nuclei including the ventral nuclei,he mediodorsal thalamic nucleus, and the pulvinar do notxpress PTH2R based on the RT-PCR and in situ hybrid-

zation histochemical data. These two techniques also pro-ided consistent data on a high level of PTH2R mRNA inhe medial geniculate body, the medial hypothalamus, andhe pontine tegmentum and a low level of PTH2R expres-ion in the lateral geniculate body, the subthalamic nu-leus, the ventral tegmental area, and the pontine reticularormation. Furthermore, neither technique reported mRNAn the substantia nigra or in the pontine nuclei. In general,

very good agreement between the results of the twoechniques confirms the findings. The only apparently con-radictory findings were that the pretectal area, where RT-CR suggested a high level of PTH2R mRNA expression,

ailed to show any labeling by using in situ hybridizationistochemistry. This difference could be the result of usingdult human brain tissue for RT-PCR and macaque brainor in situ hybridization histochemistry as overall similari-ies with small differences have also been reported be-ween the mRNA distribution of other neuropeptide sys-ems in macaque and human (Hurd et al., 1999). Theenerally good agreement between the RT-PCR and initu hybridization data, however, argues that the distribu-ion of PTH2R expression is very similar in macaque anduman, and is consistent with previous observations in rathat PTH2R expression levels do not change during post-atal development (Dobolyi et al., 2006b).

Comparison of the distribution of the PTH2R mRNA inrimates and rodents. The pattern of the PTH2R expres-ion in the brain of primates is generally similar to thatreviously reported in mice (Faber et al., 2007). The highxpression levels of the PTH2R in the septum, the caudateucleus, the medial geniculate body, the hypothalamus,he pontine tegmentum, and the cerebellum, the lowerxpression levels in the cerebral cortex, the hippocampus,he amygdala, the lateral geniculate body, the ventral teg-ental area, the pontine reticular formation, the dorsal

agal complex, and the spinal trigeminal nucleus, as wells the absence of PTH2R mRNA in most thalamic nuclei,he suprachiasmatic nucleus, the medial and lateral ma-illary nuclei, the substantia nigra, the red nucleus, therainstem motor nuclei, and most pontine and medullaryegions are all consistent with the level of PTH2R expres-

atibPtcsmtmnn2ttecr

P

mtDa2imeab2ptarcfcmbPtfp

WrtitPhhdPo

itPcdtTttspns2

TigmegantPntch

Tp

TpcTmtorwdntsbbPhiplmars

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147144

l., 2007). In situ hybridization histochemistry revealed thathe subregional distribution of PTH2R mRNA in macaques often similar to that in mice. In the medial geniculateody, the medial nucleus had the highest density ofTH2R expressing cells with some limited expression in

he ventral nucleus and no expression in the dorsal nu-leus in both macaque and mice. In the hypothalamus,imilarly high expression levels were also found in theedial preoptic area and the paraventricular and periven-

ricular nuclei in both species. Such similarities betweenRNA distributional patterns in rodents and primates areot uncommon and have also been reported for othereuropeptides and neuropeptide receptors (Elias et al.,001; Cunningham et al., 2002). Nevertheless, there werewo regions, namely the ventral part of the lateral hypo-halamus and the parabigeminal nucleus where PTH2R-xpressing cells formed compact cell groups in the ma-aque while only scattered labeled cells were found inodents, suggesting some species differences.

TH2R immunoreactivity in the primate brain

PTH2R in fiber terminals vs. cell bodies. In rat, im-unolabeling of the PTH2R with FITC-tyramide amplifica-

ion recognizes all structures described with traditionalAB visualization of immunoreactivity and also revealsdditional fine fibers not detected otherwise (Dobolyi et al.,006a). Cell bodies were not detectable by immunolabel-

ng of the PTH2R in human. This was also the case inouse, and in this species the positions of the PTH2R-xpressing cells were established by in situ hybridizations well as by the distribution of beta-galactosidase driveny the PTH2R promoter in knock-in mice (Faber et al.,007). The lack of immunolabeling within the neuronalerikarya may indicate that PTH2R protein is promptlyransported after its synthesis towards the fiber terminals,s has also been suggested in the case of several othereceptors (Deuchars et al., 2001; Stanic et al., 2006). Inontrast to most brain regions where PTH2R-ir fibers wereound together with PTH2R mRNA expressing cells, theaudate nucleus contained a significant amount PTH2RRNA in primates, as well as in mice (Faber et al., 2007)ut no PTH2R-ir fibers (Table 2) suggesting that theTH2R is expressed in neurons of the caudate nucleus

hat projects to other brain regions and that the techniquesor detection of mRNA are more sensitive than those forrotein.

Distribution of PTH2R-ir fibers and fiber terminals.e described PTH2R-ir fibers in the several different brain

egions of human and macaque including endocrine hypo-halamic structures, such as the median eminence and thenfundibular, hypothalamic paraventricular, and periven-ricular nuclei. Another brain region with a high density ofTH2R-ir fibers, the medial preoptic area, also containsypophysiotropic neurons and participates in additionalypothalamic functions, including the regulation of repro-uctive behaviors (Swann et al., 2003; Pfaff et al., 2006).TH2R-ir fibers and fiber terminals were also abundant in

ncluding the lateral septal nucleus, the paraventricularhalamic nucleus, and the periaqueductal gray. Finally,TH2R-ir fibers appeared in somato- and viscerosensoryenters, including the superficial laminae of the spinal cordorsal horn, the sensory trigeminal nuclei, the nucleus ofhe solitary tract, and the lateral parabrachial nucleus.hese topographic arrangements of PTH2R-ir fibers and

erminals in primates, like the distribution of receptor syn-hesizing cells, are very similar to that in rodents. Suchimilarities between distributional patterns in the brains ofrimates and rodents have often been reported for othereuropeptides and neuropeptide receptors suggestingimilar functions in different species (de Lacalle and Saper,000; Kostich et al., 2004).

Glutamatergic nature of PTH2R-ir fiber terminals.he quantal release of glutamate depends on its transport

nto synaptic vesicles. Therefore, the presence of vesicularlutamate transporters in nerve terminals can be used asarkers of glutamatergic presynaptic terminals (Fremeaut al., 2001). In septal and hypothalamic regions, mostlutamatergic terminals contain VGLUT2 in the rat (Lin etl., 2003). In the macaque, we observed a very denseetwork of VGLUT2-ir fiber terminals in septal and hypo-halamic regions. Some of these terminals also containedTH2R-ir and essentially all PTH2R-ir-containing termi-als in those areas contained VGLUT2-ir. The co-localiza-ion of PTH2R-ir and VGLUT2-ir suggests that PTH2R-ir-ontaining terminals in the septum, hypothalamus, and per-aps other areas are glutamatergic presynaptic terminals.

he TIP39-PTH2R neuromodulator system inrimates

IP39-expressing neurons were present in the sub-arafascicular area and in the medial paralemniscal nu-leus in 3-day-old macaque similar to the positions ofIP39-expressing cells in rodents. We did not detect TIP39RNA in human brain but the most likely explanation is

hat the level of TIP39 decreases during postnatal devel-pment in primates similar to that previously reported inodents (Dobolyi et al., 2006b; Brenner et al., 2008). Weere not able to obtain young human samples for RNAetection from the two brain regions containing TIP39eurons in rodent and monkey. Knowing that PTH2R syn-hesizing cell bodies, PTH2R containing fibers, and TIP39ynthesizing cell bodies have essentially the same distri-ution in non-human primates and rodent, that TIP39 fi-ers project to PTH2R rich regions in rodents, and thatTH2R fibers have a similar distribution in rodents anduman and non-human primates, it seems likely that there

s a similar TIP39/PTH2R match in human and non-humanrimates as well. Thus, subparafascicular TIP39 neurons

ikely project to limbic and endocrine brain regions, whileedial paralemniscal TIP39 neurons project to auditorynd viscerosensory centers, as has been demonstrated inodents (Dobolyi et al., 2003a). Furthermore, the data alsouggest that even though the human PTH2R, in contrast to

Tl

iTsaastpoadomciistrPcTwt2lti(tatms

tlmibiaD2P2thrrl2aPp

IPhhPh

AtosKLPA

B

B

B

B

C

C

C

d

D

D

D

D

D

D

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 145

IP39 (Usdin et al., 1999b), TIP39 is the physiologicaligand of the PTH2R in primates, at least in the brain.

The TIP39-PTH2R neuromodulator system may exertts action on neuronal function by presynaptic modulation.he transport of the PTH2R protein into axon terminalsupports a role of the receptor in presynaptic modulation,s previously suggested in rodents based on the presyn-ptic location of the PTH2R, as well as on the matchingubregional distribution of TIP39-ir and PTH2R-ir axonerminals (Dobolyi et al., 2006a; Faber et al., 2007). Theresynaptic modulation by TIP39 may affect the functionsf neurons that contain the PTH2R and also neurons thatre innervated by PTH2R-containing nerve terminals. Ourata provide the anatomical basis for both types of actionsf TIP39 on hypothalamic hypophysiotropic neurons. So-atostatin neurons in the hypothalamic periventricular nu-

leus project to the median eminence where somatostatins released from their terminals, reaches the anterior pitu-tary via the portal system, and inhibits growth hormoneecretion (Bluet-Pajot et al., 1998). The PTH2R content ofhese terminals provides the possibility that TIP39 directlyegulates somatostatin release by acting on theseTH2Rs. Since the activation of the PTH2R leads to in-reased cAMP and calcium levels (Behar et al., 1996)IP39 could increase somatostatin release. Such actionould be consistent with the inhibitory action of TIP39 on

he growth hormone release from the pituitary (Usdin et al.,003). TIP39 may affect CRH neurons in the paraventricu-

ar hypothalamic nucleus by a different mechanism of ac-ion. CRH neurons do not contain PTH2Rs, but they arennervated and influenced by glutamatergic synapsesCole and Sawchenko, 2002; Wittmann et al., 2005). Sincehe presynaptic terminals of some of these excitatory syn-pses contain PTH2Rs, TIP39 might increase the excita-ory influence on CRH neurons via these PTH2Rs. Such aechanism is consistent with TIP39 stimulation of CRH

ecretion from hypothalamic explants (Ward et al., 2001).Our anatomical data, together with the similarity of the

opographic distribution of the TIP39-PTH2R neuromodu-ator system in primates and rodents, suggest that in pri-

ates the TIP39-PTH2R neuromodulator system may benvolved in functional activities that have been suggestedy studies in rodents including the regulation of fear, anx-

ety, reproductive behaviors, release of pituitary hormones,nd nociceptive information processing (Ward et al., 2001;obolyi et al., 2002; Sugimura et al., 2003; Usdin et al.,003; LaBuda et al., 2004; LaBuda and Usdin, 2004;alkovits et al., 2004; Wang et al., 2006a; Fegley et al.,008). In addition, our study implies that future data ob-ained in rodents can also be realistically extrapolated touman. At present, the only human study related to a brainegion in which TIP39 neurons are concentrated is theeported increased regional blood flow during male ejacu-ation in the human subparafascicular area (Holstege et al.,003). Based on the anatomical observations in this study,dditional investigation of the functional role of the TIP39-TH2R neuromodulator system in the human brain is ex-

CONCLUSION

n conclusion, our study demonstrates the presence ofTH2R and TIP39 mRNA in the brain of human and non-uman primate, as well as PTH2R immunoreactivity in theuman CNS. The neuroanatomical distribution of theTH2R and TIP39 is quite similar in human and non-uman primates and rodents.

cknowledgments—We thank Dr. Robert Edwards for providinghe antiserum against VGLUT2. The technical assistance of Dor-ttya Kézdi and Éva Rebeka Szabó is appreciated. The work wasupported by the Hungarian Science Foundation NKTH-OTKA67646 grant for A.D., the Consortial FP6 EU Grant BrainNet IISHM-CT-2004-503039 for M.P., and the Intramural Researchrogram of the NIH, National Institute of Mental Health for T.B.U.rpád Dobolyi is grantee of the Bolyai János Scholarship.

REFERENCES

ago AG, Palkovits M, Usdin TB, Seress L, Dobolyi A (2008) Evidencefor the expression of parathyroid hormone 2 receptor in the humanbrainstem. Ideggyogy Sz 61:123–126.

ehar V, Pines M, Nakamoto C, Greenberg Z, Bisello A, Stueckle SM,Bessalle R, Usdin TB, Chorev M, Rosenblatt M, Suva LJ (1996)The human PTH2 receptor: binding and signal transduction prop-erties of the stably expressed recombinant receptor. Endocrinol-ogy 137:2748–2757.

luet-Pajot MT, Epelbaum J, Gourdji D, Hammond C, Kordon C (1998)Hypothalamic and hypophyseal regulation of growth hormone se-cretion. Cell Mol Neurobiol 18:101–123.

renner D, Bago AG, Gallatz K, Palkovits M, Usdin TB, Dobolyi A(2008) Tuberoinfundibular peptide of 39 residues in the embryonicand early postnatal rat brain. J Chem Neuroanat 36:59–68.

ole RL, Sawchenko PE (2002) Neurotransmitter regulation of cellularactivation and neuropeptide gene expression in the paraventricularnucleus of the hypothalamus. J Neurosci 22:959–969.

oolen LM, Allard J, Truitt WA, McKenna KE (2004) Central regulationof ejaculation. Physiol Behav 83:203–215.

unningham MJ, Scarlett JM, Steiner RA (2002) Cloning and distribu-tion of galanin-like peptide mRNA in the hypothalamus and pitu-itary of the macaque. Endocrinology 143:755–763.

e Lacalle S, Saper CB (2000) Calcitonin gene-related peptide-likeimmunoreactivity marks putative visceral sensory pathways in hu-man brain. Neuroscience 100:115–130.

euchars SA, Atkinson L, Brooke RE, Musa H, Milligan CJ, Batten TF,Buckley NJ, Parson SH, Deuchars J (2001) Neuronal P2X7 recep-tors are targeted to presynaptic terminals in the central and pe-ripheral nervous systems. J Neurosci 21:7143–7152.

obolyi A, Irwin S, Wang J, Usdin TB (2006a) The distribution andneurochemistry of the parathyroid hormone 2 receptor in the rathypothalamus. Neurochem Res 31:227–236.

obolyi A, Wang J, Irwin S, Usdin TB (2006b) Postnatal developmentand gender-dependent expression of TIP39 in the rat brain.J Comp Neurol 498:375–389.

obolyi A, Palkovits M, Bodnar I, Usdin TB (2003a) Neurons contain-ing tuberoinfundibular peptide of 39 residues project to limbic,endocrine, auditory and spinal areas in rat. Neuroscience 122:1093–1105.

obolyi A, Palkovits M, Usdin TB (2003b) Expression and distributionof tuberoinfundibular peptide of 39 residues in the rat centralnervous system. J Comp Neurol 455:547–566.

obolyi A, Ueda H, Uchida H, Palkovits M, Usdin TB (2002) Anatom-ical and physiological evidence for involvement of tuberoinfundibu-lar peptide of 39 residues in nociception. Proc Natl Acad Sci U S A

E

F

F

F

H

H

H

H

H

H

H

J

K

K

L

L

L

L

L

M

M

P

P

P

P

P

S

S

S

U

U

U

U

U

U

U

V

W

W

W

W

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147146

lias CF, Lee CE, Kelly JF, Ahima RS, Kuhar M, Saper CB, ElmquistJK (2001) Characterization of CART neurons in the rat and humanhypothalamus. J Comp Neurol 432:1–19.

aber CA, Dobolyi A, Sleeman M, Usdin TB (2007) Distribution oftuberoinfundibular peptide of 39 residues and its receptor, para-thyroid hormone 2 receptor, in the mouse brain. J Comp Neurol502:563–583.

egley DB, Holmes A, Riordan T, Faber CA, Weiss JR, Ma S, BatkaiS, Pacher P, Dobolyi A, Murphy A, Sleeman MW, Usdin TB (2008)Increased fear- and stress-related anxiety-like behavior in micelacking tuberoinfundibular peptide of 39 residues. Genes BrainBehav 7:933–942.

remeau RT Jr, Troyer MD, Pahner I, Nygaard GO, Tran CH, ReimerRJ, Bellocchio EE, Fortin D, Storm-Mathisen J, Edwards RH(2001) The expression of vesicular glutamate transporters definestwo classes of excitatory synapse. Neuron 31:247–260.

armar AJ (2001) Family-B G-protein-coupled receptors. GenomeBiol 2:REVIEWS3013.

artig W, Riedel A, Grosche J, Edwards RH, Fremeau RT Jr, HarkanyT, Brauer K, Arendt T (2003) Complementary distribution of vesic-ular glutamate transporters 1 and 2 in the nucleus accumbens ofrat: relationship to calretinin-containing extrinsic innervation andcalbindin-immunoreactive neurons. J Comp Neurol 465:1–10.

oare SR, Clark JA, Usdin TB (2000) Molecular determinants oftuberoinfundibular peptide of 39 residues (TIP39) selectivity for theparathyroid hormone-2 (PTH2) receptor. N-terminal truncation ofTIP39 reverses PTH2 receptor/PTH1 receptor binding selectivity.J Biol Chem 275:27274–27283.

oare SR, Usdin TB (2001) Molecular mechanisms of ligand recogni-tion by parathyroid hormone 1 (PTH1) and PTH2 receptors. CurrPharm Des 7:689–713.

olstege G, Georgiadis JR, Paans AM, Meiners LC, van der Graaf FH,Reinders AA (2003) Brain activation during human male ejacula-tion. J Neurosci 23:9185–9193.

unyady B, Krempels K, Harta G, Mezey E (1996) Immunohistochem-ical signal amplification by catalyzed reporter deposition and itsapplication in double immunostaining. J Histochem Cytochem 44:1353–1362.

urd YL, Keller E, Sotonyi P, Sedvall G (1999) Preprotachykinin-AmRNA expression in the human and monkey brain: an in situhybridization study. J Comp Neurol 411:56–72.

ohn MR, Arai M, Rubin DA, Jonsson KB, Juppner H (2002) Identifi-cation and characterization of the murine and human gene encod-ing the tuberoinfundibular peptide of 39 residues. Endocrinology143:1047–1057.

ostich WA, Grzanna R, Lu NZ, Largent BL (2004) Immunohistochem-ical visualization of corticotropin-releasing factor type 1 (CRF1)receptors in monkey brain. J Comp Neurol 478:111–125.

ronenberg HM, Lanske B, Kovacs CS, Chung UI, Lee K, Segre GV,Schipani E, Juppner H (1998) Functional analysis of the PTH/PTHrP network of ligands and receptors. Recent Prog Horm Res53:283–301.

aBuda CJ, Dobolyi A, Usdin TB (2004) Tuberoinfundibular peptide of39 residues produces anxiolytic and antidepressant actions. Neu-roreport 15:881–885.

aBuda CJ, Usdin TB (2004) Tuberoinfundibular peptide of 39 resi-dues decreases pain-related affective behavior. Neuroreport 15:1779–1782.

i C, Chen P, Smith MS (1999) Neural populations in the rat forebrainand brainstem activated by the suckling stimulus as demonstratedby cFos expression. Neuroscience 94:117–129.

in SH, Miyata S, Matsunaga W, Kawarabayashi T, Nakashima T,Kiyohara T (1998) Metabolic mapping of the brain in pregnant,parturient and lactating rats using fos immunohistochemistry. BrainRes 787:226–236.

in W, McKinney K, Liu L, Lakhlani S, Jennes L (2003) Distribution of

protein in the septum-hypothalamus of the rat. Endocrinology144:662–670.

ai JK, Assheuer J, Paxinos G (1997) Atlas of the human brain. SanDiego: Academic Press.

artin RF, Bowden DM (1996) A stereotaxic template atlas of themacaque brain for digital imaging and quantitative neuroanatomy.Neuroimage 4:119–150.

alkovits M (1973) Isolated removal of hypothalamic or other brainnuclei of the rat. Brain Res 59:449–450.

alkovits M, Dobolyi A, Helfferich F, Usdin TB (2004) Localization andchemical characterization of the audiogenic stress pathway. Ann NY Acad Sci 1018:16–24.

alkovits M, Harvey-White J, Liu J, Kovacs ZS, Bobest M, Lovas G,Bago AG, Kunos G (2008) Regional distribution and effects ofpostmortal delay on endocannabinoid content of the human brain.Neuroscience 152:1032–1039.

axinos G, Huang X (1995) Atlas of the human brainstem. Sydney:Academic Press.

faff DW, Sakuma Y, Kow LM, Lee AW, Easton A (2006) Hormonal,neural, and genomic mechanisms for female reproductive behaviors,motivation, and arousal. In: Knobil and Neill’s physiology of reproduc-tion (Neill JD, ed), pp 1825–1920. Amsterdam: Academic Press.

tanic D, Brumovsky P, Fetissov S, Shuster S, Herzog H, Hokfelt T(2006) Characterization of neuropeptide Y2 receptor protein ex-pression in the mouse brain. I. Distribution in cell bodies and nerveterminals. J Comp Neurol 499:357–390.

ugimura Y, Murase T, Ishizaki S, Tachikawa K, Arima H, Miura Y,Usdin TB, Oiso Y (2003) Centrally administered tuberoinfundibularpeptide of 39 residues inhibits arginine vasopressin release inconscious rats. Endocrinology 144:2791–2796.

wann JM, Wang J, Govek EK (2003) The MPN mag: introducing acritical area mediating pheromonal and hormonal regulation ofmale sexual behavior. Ann N Y Acad Sci 1007:199–210.

sdin TB (1997) Evidence for a parathyroid hormone-2 receptor se-lective ligand in the hypothalamus. Endocrinology 138:831–834.

sdin TB, Bonner TI, Hoare SR (2002) The parathyroid hormone 2(PTH2) receptor. Receptors Channels 8:211–218.

sdin TB, Dobolyi A, Ueda H, Palkovits M (2003) Emerging functionsfor tuberoinfundibular peptide of 39 residues. Trends EndocrinolMetab 14:14–19.

sdin TB, Gruber C, Bonner TI (1995) Identification and functionalexpression of a receptor selectively recognizing parathyroid hor-mone, the PTH2 receptor. J Biol Chem 270:15455–15458.

sdin TB, Hilton J, Vertesi T, Harta G, Segre G, Mezey E (1999a)Distribution of the parathyroid hormone 2 receptor in rat: immuno-localization reveals expression by several endocrine cells. Endo-crinology 140:3363–3371.

sdin TB, Hoare SR, Wang T, Mezey E, Kowalak JA (1999b) TIP39:a new neuropeptide and PTH2-receptor agonist from hypothala-mus. Nat Neurosci 2:941–943.

sdin TB, Wang T, Hoare SR, Mezey E, Palkovits M (2000) Newmembers of the parathyroid hormone/parathyroid hormone recep-tor family: the parathyroid hormone 2 receptor and tuberoinfun-dibular peptide of 39 residues. Front Neuroendocrinol 21:349–383.

arga T, Palkovits M, Usdin TB, Dobolyi A (2008) The medialparalemniscal nucleus and its afferent neuronal connections in rat.J Comp Neurol 511:221–237.

ang J, Coolen LM, Brown JL, Usdin TB (2006a) Neurons containingtuberoinfundibular peptide of 39 residues are activated followingmale sexual behavior. Neuropeptides 40:403–408.

ang J, Palkovits M, Usdin TB, Dobolyi A (2006b) Afferent connectionsof the subparafascicular area in rat. Neuroscience 138:197–220.

ang J, Palkovits M, Usdin TB, Dobolyi A (2006c) Forebrain projec-tions of tuberoinfundibular peptide of 39 residues (TIP39)-contain-ing subparafascicular neurons. Neuroscience 138:1245–1263.

ang T, Palkovits M, Rusnak M, Mezey E, Usdin TB (2000) Distributionof parathyroid hormone-2 receptor-like immunoreactivity and mes-

W W

A. G. Bagó et al. / Neuroscience 162 (2009) 128–147 147

ard HL, Small CJ, Murphy KG, Kennedy AR, Ghatei MA,Bloom SR (2001) The actions of tuberoinfundibular peptideon the hypothalamo-pituitary axes. Endocrinology 142:3451–

ittmann G, Lechan RM, Liposits Z, Fekete C (2005) Glutamatergicinnervation of corticotropin-releasing hormone- and thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic para-

(Accepted 22 April 2009)(Available online 3 May 2009)