Chromaffin Cell Function and Structure Is Impaired in Corticotropin

Home , Chromaffin cell

Molecular Psychiatry (2002) 7, 967–974  2002 Nature Publishing Group All rights reserved 1359-4184/02 $25.00 www.nature.com/mp ORIGINAL RESEARCH ARTICLE Chromafﬁn cell function and structure is impaired in corticotropin-releasing hormone receptor type 1-null mice M Yoshida-Hiroi1,2, MJ Bradbury3, G Eisenhofer2, N Hiroi1,5, WW Vale3, GE Novotny4, HG Hartwig5, WA Scherbaum6 and SR Bornstein6

1Pediatric and Reproductive Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA; 2Clinical Neurocardiology Section and Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA; 3The Clayton Foundation Laboratories for Peptide Biology, The Salk Institute, La Jolla, CA, USA; 4Institute of Neuroanatomy, Heinrich-Heine University, Du¨sseldorf, Germany; 5Department of Anatomy, Heinrich-Heine University, Du¨sseldorf, Germany; 6Department of Endocrinology, Heinrich-Heine University, Du¨sseldorf, Germany

Corticotropin-releasing hormone (CRH) is both a main regulator of the hypothalamic-pituitary- adrenocortical axis and the autonomic nervous system. CRH receptor type 1 (CRHR1)- deficient mice demonstrate alterations in behavior, impaired stress responses with adrenocortical insufficiency and aberrant neuroendocrine development, but the adrenal medulla has not been analyzed in these animals. Therefore we studied the production of adrenal catecholamines, expression of the enzyme responsible for catecholamine biosynthesis neuropeptides and the ultrastructure of chromaffin cells in CRHR1 null mice. In addition we examined whether treatment of CRHR1 null mice with adrenocorticotropic hormone (ACTH) could restore function of the adrenal medulla. CRHR1 null mice received saline or ACTH, and wild- type or heterozygous mice injected with saline served as controls. Adrenal epinephrine levels in saline-treated CRHR1 null mice were 44% those of controls (P Ͻ0.001), and the phenylethanolamine N-methyltransferase (PNMT) mRNA levels in CRHR1 null mice were only 25% of controls (P Ͻ0.001). ACTH treatment increased epinephrine and PNMT mRNA level in CRHR1 null mice but failed to restore them to normal levels. Proenkephalin mRNA in both saline- and ACTH-treated CRHR1 null mice were higher than in control animals (215.8% P Ͻ0.05, 268.9% P Ͻ0.01) whereas expression of neuropeptide Y and chromogranin B did not differ. On the ultrastructural level, chromaffin cells in saline-treated CRHR1 null mice exhibited a marked depletion in epinephrine-storing secretory granules that was not completely normalized by ACTH-treatment. In conclusion, CRHR1 is required for a normal chromaffin cell structure and function and deletion of this gene is associated with a significant impairment of epinephrine biosynthesis. Molecular Psychiatry (2002) 7, 967–974. doi:10.1038/sj.mp.4001143 Keywords: CRH receptor type 1; chromaffin cell; catecholamine; ACTH; adrenal; mouse

Introduction through two G-protein-coupled receptors, CRHR1 and CRHR2, with CRHR1 being the major receptor Corticotropin-releasing hormone (CRH) is a major coor- mediating the activation of the HPA axis in rodents. dinator of the hypothalamic-pituitary-adrenocortical Genetic manipulation of each element of the CRH-sys- (HPA) axis.1 CRH released from the paraventricular tem is associated with various degrees of impairment nucleus (PVN) of the hypothalamus activates CRH of adrenocortical structure and function.3–11 receptors on anterior pituitary corticotropes, resulting CRH-deficient mice lack the normal diurnal gluco- in secretion of adrenocorticotropic hormone (ACTH) corticoid rhythm and present with adrenocortical atro- which in turn activates the synthesis and release of glu- phy and severely impaired glucocorticoid release to the cocorticoids from the adrenal cortex.1,2 CRH acts stress of restraint, ether and fasting.4 Similarly, CRHR1 null mice demonstrate adrenal atrophy and a reduced Correspondence: M Yoshida-Hiroi, Clinical Neurocardiology Sec- stress-induced secretion of corticosterone.6,7 CRHR2 tion and Surgical Neurology Branch, National Institute of Neuro- null mice present with normal corticosterone secretion logical Disorders and Stroke, National Institutes of Health, Bldg to acute stress but show early termination of ACTH 10, Room 6N252, 10 Center Drive, Bethesda, Maryland 20892- 1620, USA. E-mail: hiroimȰninds.nih.gov release and an altered recovery phase of the HPA 8 Received 22 October 2001; revised 18 February 2002; accepted reponse with elevated corticosterone levels. 21 February 2002 There is a functional interdependence between the Chromaffin cell function in CRHR1-null mice M Yoshida-Hiroi et al 968 two endocrine cell systems in the adrenal gland.12 measurements because data are expressed per gland. While catecholamines and neuropeptides produced in The left adrenals from each pair were immediately the adrenal medulla regulate adrenocortical function, snap frozen and stored at −80°C until homogenized. adrenocortical steroids influence the structure and The right adrenals were fixed for microscopy. function of the chromaffin cells in the adrenal medulla.2,13,14 Glucocorticoids induce the enzyme respon- Assay of tissue catecholamines sible for catecholamine biosynthesis and glucocort- Adrenal concentrations of catecholamines (norepine- icoid receptor-deficient mice have no adrenal phrine, epinephrine, dopamine, and dihydroxy- epinephrine production.15 Likewise, patients with phenylalanine) were quantified by high-performance adrenocortical insufficiency due to Addison’s disease liquid chromatography (HPLC) with electrochemical or 21-OH deficiency have low plasma epinephrine lev- detection.22 Adrenals were homogenized in 500 ␮lof els and adrenomedullary dysplasia.16–18 Recent studies cold 0.4 M perchloric acid containing 0.5 mM EDTA. demonstrated low epinephrine levels and a decrease in Homogenized samples were centrifuged (3000 rpm for adrenal PNMT expression in CRH-deficient mice4,6,7 30 min at 4°C) and supernatants collected and stored and adrenomedullary atrophy has been reported in at −80°C until assayed. Concentrations of catecholam- CRHR1 null mice.6 ines in supernatants were determined after extraction CRHR1 receptors are expressed on chromaffin cells19 using an alumina adsorption method described pre- and a direct action of CRH on chromaffin cell function viously.22 has been described.20,21 CRHR1 null mice are an inter- esting model to analyze the role of the CRH system for Total RNA isolation and TaqMan PCR the functioning of the adrenal medulla. Therefore we Total RNA was isolated from the adrenal glands of studied the production of adrenal hormones, CRHR1 null and wild-type mice using the microRNA expression of adrenal enzymes, StAR, PNMT, neuro- isolation kit (Stratagene, La Jolla, CA, USA). RT-PCR peptides and the ultrastructure of adrenal cells in experiments were carried out according to the THER- CRHR1-deficient mice. In addition we examined MOSCRIPT RT-PCR system kit (GIBCO, Gaithersburg, whether treatment of CRHR1 null mice with ACTH MD, USA) after treatment with deoxyribonuclease I could restore the adrenal structure and function. (GenHunter Corp, Nashville, TN, USA). Total RNA (1 ␮g) of adrenal glands of each group were reverse transcribed to complementary DNA (cDNA) by a reac- Materials and methods tion containing 2 mM deoxynucleotide mix, 100 mM Mice DTT, 40 units RNase inhibitor, 50 ng random primer The mice in which exons 5–8 of the CRHR1 were and 15 units thermoscript reverse transcriptase. The replaced with a PCK-neomycin-resistant cassette reaction was run at 25°C for 10 min and 50°C for (CRHR1 null)7 and their wild-type littermates used in 50 min, heated at 85°C for 5 min, and then cooled to these experiments were bred from a colony at the Salk 4°C. Institute for Biological Research. The CRHR1 geno- To quantify the expression of phenylethanolamine types were identified by PCR analysis using DNA iso- N-methyltransferase (PNMT), steroidogenic acute regulated from tail samples.7 All animals were bred under latory protein (StAR), neuropeptide Y (NPY), chromo- pathogen-free conditions, fed on a normal diet and granin B and proenkephalin, we applied the TaqMan maintained in the transgenic animal facility of the Salk PCR using the 7700 Sequence Detector (Perkin-Elmer Institute. Maintenance and experiments were conduc- Applied Biosystems, Foster, CA, USA) as described ted according to the Institutional Animal Care and Use previously.23 Reactions contained 1× TaqMan Univer- committee at the Salk Institute for Biological Research, sal PCR Master Mix, 900 nM of forward and reverse La Jolla, CA, USA. primers for PNMT, StAR, NPY, chromogranin B or pro- Mice were weaned at 21 days of age. Starting at enkephalin, and 200 nM of TaqMan probe (Table 1). weaning, seven CRHR1 null mice received normal Thermal cycling proceeded with 40 cycles of 95°C for saline (50 ␮l) and six null mice were treated with 0.4 U 15 s and 60°C for 1 min. Input RNA amounts were cal- of ACTH (ACTHar gel, Rhone-Poulenc Rorer, Col- culated with multiplex comparative method for both legeville, PA, USA). Mice were injected subcutane- the mRNAs of interest (PNMT, StAR, NPY, chromo- ously at 4 pm, 2 h before lights turned off at 6 pm dur- granin B or proenkephalin) and 18S. ing a 12:12 h light cycle. Ten wild-type or heterozygous mice injected with normal saline served as controls. Histology and electron microscopy Fourteen days later, mice were restrained for up to The adrenals were embedded in tissue tek. Five mm 4 min while blood was obtained through the retroorbi- sections were mounted on poly-L-Lysin (Sigma, Mun- tal sinus. Serum was separated from blood samples and ich, Germany) coated slides. After air drying, sections assayed for corticosterone (ICN Diagnostics, Costa were fixed in acetone for 15 min. The sections were Mesa, CA, USA). Thymus glands were dissected and stained with hematoxylin-eosin and screened for infil- weighed. After killing, adrenal glands for PCR were tration. carefully removed with surrounding fat with microfor- For ultrastructural investigations, samples of tissue ceps and cleaned of extraneous material. The residual were fixed with 2% formaldehyde and 2% glutaral- fat and tissue is not a confound for the subsequent dehyde. Thereafter, the samples were incubated in

Molecular Psychiatry Chromafﬁn cell function in CRHR1-null mice M Yoshida-Hiroi et al 969 Table 1 Sequences of primers and TaqMan probes

Primer or probe GenBank Direction Sequence 5Ј–3Ј accession No.

Phenylethanolamine L12678 Forward GTC GGG ACG GGT TCT CAT T N-methyltransferase Reverse CCA AGA AGT CTG TCA TGG TGA TG Probe FAM-CTC CGG CCC CAC CAT ATA TCA GCT G-TAMRA Steroidogenic acute AY032730 Forward CCC ATG GAC AGA CTC TAT GAA GAA regulatory protein Reverse TGA CAT TTG GGT TCC ACT CTC C Probe FAM-TTG TGG ACC GCA TGG AGG CCA-TAMRA Neuropeptide Y NM012614 Forward TGT GGA CTG ACC CTC GCT CTA T Reverse AGC GGA GTA GTA TCT GGC CAT GT Probe FAM-CTC GTG TGT TTG GGC ATT CTG GCT G-TAMRA Chromogranin B X53028 Forward CCC GCT GGC TGA ACT TTT C Reverse GAG TTC TGA CGG CGG AAG AG Probe FAM-CAG ATC CAT CGC AGC CAA GTT CTC CA-TAMRA Proenkephalin M13227 Forward AGA AGC GAA CGG AGG AGA GAT Reverse TCC CGT TCC CAG TAG CTC TTT Probe FAM-TTG GCC AAC TCC TCC GAT CTG CTG-TAMRA

0.1 M phosphate buffer at pH 7.3 for 3 h. Samples were Adrenal catecholamine levels postfixed for 90 min (2% OsO4 in 0.1 M cacodylate Adrenal epinephrine levels were 56% lower in saline- buffer, pH 7.3), dehydrated in ethanol and embedded treated CRHR1 null mice (1318.5 ± 172.5 ng adrenal in epoxy resin. Ultrathin sections were mounted on gland−1; AD, P Ͻ0.001) than controls (3025.4 ± 234.2 ng 200-mesh uncoated nickel grids. Sections were stained AD−1) (Figure 1). Adrenal epinephrine levels in ACTH- with uranyl acetate and lead Na citrate for 24 h and treated CRHR1 null mice were increased (2374.6 ± examined under a HITACHI electron microscope H- 209.1 ng AD−1)(P Ͻ0.05) to levels greater than in 600 (Hitachi, Japan) and photographed. saline-treated CRHR1 null mice, but were not fully restored to the level in control mice. Adrenal norepi- Statistical analysis nephrine levels among the three groups were not dif- ferent (1129.5 ± 112.5, 1166.9 ± 150.2, 1415.2 ± 66.0 ng Data analysis was performed by one-way ANOVA and −1 Fisher’s Protected Least Significant Difference for cat- AD in saline-treated CRHR1 null mice, ACTH-treated echolamine. Expressions of mRNA were analyzed CRHR1 null mice, and controls, respectively). using Kruskal–Wallis test. Results are expressed as the mean ± SEM. Statistical significance was defined as P Expression of PNMT, StAR, NPY, chromogranin B Ͻ0.05. and proenkephalin mRNA in adrenal glands Expression of PNMT mRNA in adrenals of saline- treated CRHR1 null mice was markedly reduced com- Results Hormonal analysis Serum corticosterone concentrations, 2 h after lights on, in saline-treated CRHR1 null mice were significantly lower than those of control mice (6.1 ± 3.6 ng ml−1 vs 46.4 ± 13.3 ng ml−1, P Ͻ0.01). Corticosterone concentrations in ACTH-treated CRHR1 null mice were equally low (8.3 ± 4.5 ng ml−1), likely due to the 16 h interval between injection and sampling. Nonetheless, ACTH treatment appeared to provide a physiological adrenal stimulation during the dark phase of the light cycle. As expected in corticosteroid-deficient animals, thymus weights in saline-treated CRHR1-null mice Figure 1 Concentrations of adrenal catecholamines in were significantly larger than those in controls (0.079 CRHR1 null mice and control mice. Mean Ϯ SEM, asterisks: ± 0.006 vs 0.05 ± 0.014 g, P Ͻ0.01). ACTH-treatment statistically significant differences between concentrations of of the CRHR1-null mice restored this glucocorticoid- CRHR1 null mice and controls,* P Ͻ0.05, **P Ͻ0.01, ***P sensitive tissue (0.058 ± 0.0042 g). Ͻ0.001; EPI, epinephrine; NE, norepinephrine.

Molecular Psychiatry Chromafﬁn cell function in CRHR1-null mice M Yoshida-Hiroi et al 970 pared to controls (22.5 ± 3.1% reduction of control, P Ͻ0.01) (Figure 2). PNMT expression levels in ACTH- treated CRHR1 null mice were 42.7 ± 9.1% less than those in controls (P Ͻ0.05). Expression of StAR mRNA in CRHR1 null mice was distinctly decreased less than controls (28.0 ± 6.4% reduction of control, P Ͻ0.001). StAR expression levels in ACTH-treated CRHR1 null mice was blunted more than saline-treated CRHR1 null mice but clearly less than that control (42.1 ± 9.0% reduction of control, P Ͻ0.01). NPY and chromogranin B mRNA expression did not differ among the three groups. The level of proenkephalin in both saline and ACTH-treated CRHR1 null mice was increased compared to controls (215.8 ± 31.9%, P Ͻ0.05; 268.9 ± 22.3%, P Ͻ0.01).

Ultrastructural analysis of the adrenal glands Adrenocortical cells of control mice demonstrated irregularly shaped nuclei and the characteristic mitochondrial structures with tubulovesicular cristae and ample smooth endoplasmic reticulum (SER). Adreno- cortical cells contained few liposomes and exhibited some ﬁlopodia on the cell surface. (Figure 3a). Saline- treated CRHR1 null mice displayed an increase in liposomes compared with controls (Figure 3b). The amount of SER and the number of mitochondria were reduced, consistent with decreased corticosterone production in these animals. Also, the internal mitochondrial membranes were reduced and exhibited a more tubular arrangement characteristic of adrenocortical cells in a hypofunctional state. Following ACTH injections, adrenocortical structure and ultrastructure were restored (Figure 3c). Cortical cells demonstrated an increase in ﬁlopodia, SER and mitochondria. The mitochondrial membranes displayed a dense vesicular pat-

Figure 3 Adrenocortical ultrastructure of CRHR1 null mice (+/Ϫ ACTH) and controls. In wild-type animals adrenocortical cells exhibit numerous round vesicular mitochondria (Mit) and filopodia (arrows) but few liposomes (Lip) (a). In Figure 2 Expression of PNMT, StAR, NPY, chromogranin B contrast in CRHR1 null mice the number of lipid droplets and proenkephalin mRNA in adrenal grands. The expression are increased while the amount of mitochondria and smooth of PNMT, StAR, NPY, chromogranin B and proenkephalin endoplasmic reticulum (Ser) is reduced (b). ACTH treatment mRNA were determined by quantitative TaqMan PCR in the restores adrenocortical cell morphology with a marked adrenal grands of CRHR1 null mice and controls. Results are increase in mitochondria and the cell surface presents numer- standardized to expression of controls mRNA (100%). Mean ous filopodia. Liposomes are reduced (c). Nuc ϭ nucleus, bar Ϯ SEM, asterisks: statistically significant differences between ϭ 1 ␮m, cells stained with uranyl acetate in all three panels. expressions of CRHR1 null mice and controls, *P Ͻ0.05, **P Ͻ0.01, ***P Ͻ0.001; Chro B, chromogranin B; proE, proenkephalin.

Molecular Psychiatry Chromaffin cell function in CRHR1-null mice M Yoshida-Hiroi et al 971 tern, reflecting an adequate trophic effect of ACTH on the adrenal cortex. Adrenomedullary cells in controls had the characteristic ultrastructural features of neuroendocrine cells with an ample presence of membrane bound, secretory ranging from 60 to 400 nm in greatest dimension, and rough endoplasmic reticulum (RER) (Figure 4a). The two principal types of chromaffin vesicles were epinephrine and norepinephrine. Epinephrine-containing vesicles were large, round or elongated medium-density. Norepinephrine-containing vesicles were small and electron-dense. In controls the cytoplasm was filled with both types of secretory granules. In contrast, in chromaffin cells in saline-treated CRHR1 null mice, the number of secretory granules was clearly reduced. The remaining granules were smaller and primarily electron dense (Figure 4b). Adrenomedullary cells in ACTH-treated CRHR1 null mice displayed an increase in medium-density secretory vesicles and RER but normal chromaffin structure was not restored (Figure 4c).

Discussion CRH is a crucial regulator of the neuroendocrine, autonomic and behavioral response to stress.1,24–26 In the present study we demonstrate a significant impairment of chromaffin cell function and structure in CRHR1- deficient mice. In accordance with a reduced epinephrine production there was a decline in PNMT expression and depletion in epinephrine storing secretory granules in chromaffin cells. ACTH injections at physiologically relevant doses in the CRHR1 null mice only partially restored chromaffin cell structure and function to that observed in controls. CRHR1 deletion and ACTH replacement had expected effects on the ultrastructural parameters of the adrenal cortex. CRHR1 null animals demonstrated a reduced number of internal mitochondrial membranes and SER with an increase in liposomes. There is a well-established correlation between the production sites for steroidogenesis in adrenocortical cells, as indi- cated by the number and density of inner mitochondrial membranes and SER, and the biochemical activity of these cells.27,28 Likewise, the increase in liposomes, the storage organelles for cholesterol, reflect reduced steroid biosynthesis and a hypofunctional adrenocortical state.29 Consistent with these assumptions, ACTH injections at a dose sufficient to restore thymus weight Figure 4 Adrenomedullary ultrastructure of CRHR1-null + Ϫ induced a strong increase in the number of mitochon- mice ( / ACTH) and controls. In wild-type animals the cyto- dria and increased density of their internal membranes plasm of chromaffin cells is filled with chromaffin vesicles (CV) of both granular/epinephrine and electron-dense/ while the number of liposomes were markedly norepinephrine storing secretory granules 50–450 nm in reduced. This was associated with a significant diameter. Cells contain rough endoplasmic reticulum (Rer) increase in adrenal epinephrine production. However, and cristae-like mitochondria (Mit) (a). In the CRHR1 null ACTH failed to fully restore corticosterone level, StAR mice there is a reduction in chromaffin vesicles with the expression, chromaffin cell structure and PNMT remaining granules presenting as electron dense (b). ACTH expression in these animals. The mechanism behind treatment fails to restore normal chromaffin ultrastructure in this disparity of ACTH replacement effects reflects the the CRHR1 null mice (c). In all three panels the bar represents complex regulation of chromaffin cell functions. 1 ␮m. Adequate intraadrenal concentrations of catecholamines are required for full chromaffin cell activity. For example, steroidogenic factor 1 (SF-1) heterozygous

Molecular Psychiatry Chromaffin cell function in CRHR1-null mice M Yoshida-Hiroi et al 972 mice have impaired-induced glucocorticoid pro- mixed results, with some groups reporting decreases duction coupled with a significant defect in adreno- in proenkephalin mRNA, and others an increase after medullary development and catecholamine pro- incubation with reserpine and/or tetrabenzamine.45–48 duction.30 Therefore, this increase in neuropeptide production The use of non-peptide CRHR1 antagonist, antalar- may be the consequence of an increased pre-ganglionic min reduces plasma Epi and NE responses to stress in stimulation of the chromaffin cells in CRHR1 null primates.31 In addition, antalarmin blocks the central mice. Interestingly, these results are consistent with CRHR1-mediated hypertension in rats.32 Epi levels tyrosine hydroxylase-deficient mice that have blunted were decreased in CRHR1 null mice. NE levels did not catecholamine secretion and low corticosterone pro- change compared to controls in this study. The cause duction.49 The expression of chromogranin B, a major of the difference in NE levels between the two studies vesicle compound in the adrenal medulla of mice, and is unclear. However, CRH null mice have a low Epi NPY were not significantly altered in the CRHR1 null level associated with decreased expression of PNMT in mice. Chromogranin B levels have been shown to be the adrenal medulla.33 Thus, the cause may reflect a unaffected by hypophysectomy and consequently low species difference. Glucocorticoids exert marked and corticosterone levels and were not significantly affec- complex influences on PNMT gene transcription. For ted by increased neuronal activity generated with insu- example, stimulation of the PNMT gene by the neural lin treatment.44 This concurs with our findings in the crest factor activating protein-2 (AP-2) requires a direct CRHR1 null mice and confirms the notion that differ- interaction with ligand-activated type II glucocorticoid ent classes of secretory proteins presented in chromaf- receptors.34 In addition, the PNMT gene contains over- fin vesicles are regulated in a differential manner in lapping consensus elements for the promoter selective vivo. transcription factor (Sp 1) and the immediate early The finding of impaired adrenal catecholamine and gene transcription factor Egr-1 that are capable of dif- neuropeptide production in CRHR1 null mice should ferentially activating PNMT gene expression.35 Sp 1, be considered when interpreting the behavioral abnor- Egr-1 and the glucocorticoid receptor and AP-2 func- malities in these animals. We have recently reported tion cooperatively to stimulate PNMT promoter transgenic animals overexpressing PNMT and high epi- activity through poorly delineated mechanisms.36 nephrine levels in the brain.50 These animals had meta- Glucocorticoid control of PNMT gene transcription bolic alterations and were highly aggressive, some- and protein synthesis do not fully account for changes times escalating in the killing of caged males.51 On the in PNMT expression in the present study. The capacity other hand, epinephrine-deficient animals created by of ACTH injections to increase epinephrine biosyn- deletion of the enzyme responsible for the conversion thesis while PNMT expression remains markedly of norepinephrine to epinephrine (PNMT) did not decreased suggests that corticosteroids can posttran- exhibit major changes of viability, metabolism, or scriptionally regulate PNMT protein expression.37 Alt- behavior.52 ered PNMT expression in CRHR1 null mice may reflect In conclusion, CRHR1 is required for normal chrom- direct effects of gene deletion in adrenal chromaffin affin cell development and function and deletion of cells. Chromaffin cells express CRHR1 receptors and this gene is associated with a significant impairment of there is an intraadrenal CRH/ACTH system under feed- epinephrine biosynthesis. Novel preclinical and clini- back control of glucocorticoids.12,19 Exogenous CRH cal strategies employing non-peptidic antagonists for preserves chromaffin cell morphology in vitro and CRHR1 should keep in mind the functional interde- stimulates catecholamine production in dispersed rat pendence of the endocrine stress response in vivo. chromaffin and human pheochromocytoma cells.21 Hypoxia-induced adrenal chromaffin cell changes are References partially reversed with exogenous CRH. Thus deficits in chromaffin cell in saline and ACTH-treated CRHR1 1 Chrousos GP, Gold PW. The concepts of stress and stress system disorders. Overview of physical and behavioral homeostasis. JAMA null mice may reflect inadequate stimulation of the 1992; 267: 1244–1252. adrenal CRHR1. 2 Bornstein SR, Chrousos GP. Clinical Review: adrenocorticotropin To further analyze the functional state of the chrom- (ACTH)- and non-ACTH mediated regulation of the adrenal cortex. affin cells in CRHR1-null mice, we performed a quanti- Neural and immune inputs. J Clin Endocrinol Metab 1999; 84: 1729–1736. tative mRNA analysis of proenkephalin, NPY and chro- 3 Muglia LJ, Jacobson L, Weninger SC, Luedeke CE, Bae DS, Jeong mogranin B. Proenkephalin, NPY and chromogranins KH, Majzoub JA. Impaired diurnal adrenal rhythmicity restored by were chosen because they are major components stored constant infusion of corticotropin-releasing hormone in corticotro- with catecholamines in chromaffin cells.38–41 While pin-releasing hormone-deficient mice. J Clin Invest 1997; 99: met-enkephalin constitutes the most abundant neuro- 2923–2929. 4 Muglia L, Jacobson L, Dikkes P, Majzoub JA. Corticotropin-releas- peptide produced by chromaffin cells and correlates ing hormone deficiency reveals major fetal but not adult glucocort- with catecholamines, NPY is regulated in a differential icoid need. Nature 1995; 373:427–432. manner, following neural activation.42–44 In ACTH and 5 Coste SC, Kesterson RA, Heldwein KA, Stevens SL, Heard AD, saline-treated CRHR1 null mice, there was an increase Hollis JH et al. Abnormal adaptations to stress and impaired car- diovascular function in mice lacking corticotropin-releasing hor- in pro-enkephalin RNA levels. Studies that examined mone receptor-2. Nat Genet 2000; 24:403–409. the level of regulation of proenkephalin synthesis by 6 Timpl P, Spanagel R, Sillaber I, Kresse A, Reul JM, Stalla GK et al. catecholamine-depleting agents in vitro have produced Impaired stress response and reduced anxiety in mice lacking a

Molecular Psychiatry Chromaffin cell function in CRHR1-null mice M Yoshida-Hiroi et al 973 functional corticotropin-releasing hormone receptor. Nat Genet 28 Nussdorfer GG. Cytophysiology of the adrenal cortex. Int Rev Cytol 1998; 19:162–166. 1986; 98:1–405. 7 Smith GW, Aubry JM, Dellu F, Contarino A, Bilezikjian LM, Gold 29 Bornstein SR, Ehrhart-Bornstein M, Scherbaum WA. Morphologi- LH et al. Corticotropin releasing factor receptor 1-deficient mice cal and functional studies of the paracrine interaction between cor- display decreased anxiety, impaired stress response, and aberrant tex and medulla in the adrenal gland. Microsc Res Tech 1997; 36: neuroendocrine development. Neuron 1998; 20: 1093–1102. 520–533. 8 Bale TL, Contarino A, Smith GW, Chan R, Gold LH, Sawchenko PE 30 Bland ML, Jamieson CA, Akana SF, Bornstein SR, Eisenhofer G, et al. Mice deficient for corticotropin-releasing hormone receptor-2 Dallman MF, Ingraham HA. Haploinsufficiency of steroidogenic display anxiety-like behaviour and are hypersensitive to stress. Nat factor-1 in mice disrupts adrenal development leading to an Genet 2000; 24: 410–414. impaired stress response. Proc Natl Acad Sci USA 2000; 97: 9 Stenzel-Poore MP, Duncan JE, Rittenberg MB, Bakke AC, Heinrichs 14488–14493. SC. CRH overproduction in transgenic mice: behavioral and 31 Habib KE, Weld KP, Rice KC, Pushkas J, Champoux M, Listwak immune system modulation. Ann N Y Acad Sci 1996; 780:36–48. S et al. Oral administration of a corticotropin-releasing hormone 10 Stenzel-Poore MP, Cameron VA, Vaughan J, Sawchenko PE, Vale receptor antagonist significantly attenuates behavioral, neuroen- W. Development of Cushing’s syndrome in corticotropin-releasing docrine, and autonomic responses to stress in primates. Proc Natl factor transgenic mice. Endocrinology 1992; 130: 3378–3386. Acad Sci USA 2000; 97:6079–6084. 11 Burrows HL, Nakajima M, Lesh JS, Goosens KA, Samuelson LC, 32 Briscoe RJ, Cabrera CL, Baird TJ, Rice KC, Woods JH. Antalarmin Inui A et al. Excess corticotropin releasing hormone-binding pro- blockade of corticotropin releasing hormone-induced hypertension tein in the hypothalamic-pituitary-adrenal axis in transgenic mice. in rats. Brain Res 2000; 881: 204–207. J Clin Invest 1998; 101: 1439–1447. 33 Karalis KP, Kontopoulos E, Muglia LJ, Majzoub JA. Corticotropine- 12 Ehrhart-Bornstein M, Hinson JP, Bornstein SR, Scherbaum WA, releasing hormone deficiency unmasks the proinflammatory effect of epinephrine. Proc Natl Acad Sci USA 1999; 96: 7093–7097. Vinson GP. Intraadrenal interactions in the regulation of adreno- 34 Ebert SN, Ficklin MB, Her S, Siddall BJ, Bell RA, Ganguly K et al. cortical steroidogenesis. Endocr Rev 1998; 19:101–143. Glucocorticoid-dependent action of neural crest factor AP-2: stimu- 13 Axelrod J, Reisine TD. Stress hormones: their interaction and regulation of phenylethanolamine N-methyltransferase gene lation. Science 1984; 224:452–459. expression. J Neurochem 1998; 70: 2286–2295. 14 Wurtman RJ, Axelrod J. Control of enzymatic synthesis of adre- 35 Ebert SN, Wong DL. Differential activation of the rat phenylethano- naline in the adrenal medulla by adrenal cortical steroids. J Biol lamine N-methyltransferase gene by Sp1 and Egr-1. J Biol Chem Chem 1966; 241: 2301–2305. 1995; 270: 17299–17305. 15 Cole TJ, Blendy JA, Monaghan AP, Krieglstein K, Schmid W, 36 Wong DL, Siddall BJ, Ebert SN, Bell RA, Her S. Phenylethanolam- Aguzzi A et al. Targeted disruption of the glucocorticoid receptor ine N-methyltransferase gene expression: synergistic activation by gene blocks adrenergic chromaffin cell development and severely Egr-1, AP-2 and the glucocorticoid receptor. Brain Res Mol Brain retards lung maturation. Genes Dev 1995; 9: 1608–1621. Res 1998; 61:154–161. 16 Bornstein SR, Tajima T, Eisenhofer G, Haidan A, Aguilera G. Adre- 37 Wong DL, Siddall B, Wang W. Hormonal control of rat adrenal nomedullary function is severely impaired in 21-hydroxylase- phenylethanolamine N-methyltransferase. Enzyme activity, the deficient mice. FASEB J 1999; 13: 1185–1194. final critical pathway. Neuropsychopharmacology 1995; 13: 223– 17 Merke DP, Chrousos GP, Eisenhofer G, Weise M, Keil MF, Rogol 234. AD et al. Adrenomedullary dysplasia and hypofunction in patients 38 Viveros OH, Diliberto EJ Jr, Hong JH, Kizer JS, Unsworth CD, Kana- with classic 21-hydroxylase deficiency. N Engl J Med 2000; 343: matsu T. The regulation of enkephalin levels in adrenomedullary 1362–1368. cells and its relation to chromaffin vesicle biogenesis and func- 18 Bornstein SR, Breidert M, Ehrhart-Bornstein M, Kloos B, Scher- tional plasticity. Ann N Y Acad Sci 1987; 493: 324–341. baum WA. Plasma catecholamines in patients with Addison’s dis- 39 O’Connor DT, Klein RL, Thureson-Klein AK, Barbosa JA. Chromo- ease. Clin Endocrinol (Oxf) 1995; 42: 215–218. granin A: localization and stoichiometry in large dense core cat- 19 Willenberg HS, Bornstein SR, Hiroi N, Path G, Goretzki PE, Scher- echolamine storage vesicles from sympathetic nerve. Brain Res baum WA, Chrousos GP. Effects of a novel corticotropin-releasing- 1991; 567:188–196. hormone receptor type I antagonist on human adrenal function. 40 Dillen L, Miserez B, Claeys M, Aunis D, De Potter W. Posttrans- Mol Psychiatry 2000; 5:137–141. lational processing of proenkephalins and chromogranins/ 20 Hoheisel G, Schauer J, Scherbaum WA, Bornstein SR. The effect of secretogranins. Neurochem Int 1993; 22: 315–352. corticotropin-releasing hormone (CRH) on the adrenal medulla in 41 Tran MA, Damase-Michel C, Tavernier G, Giraud P, Montastruc JL, hypophysectomized rats. Histol Histopathol 1998; 13:81–87. Montastruc P. Co-release of neuropeptides and catecholamines by 21 Venihaki M, Gravanis A, Margioris AN. Comparative study adrenal medulla. Arch Mal Coeur Vaiss 1993; 86:1253–1256. between normal rat chromaffin and PC12 rat pheochromocytoma 42 Bastiaensen E, De Block J, De Potter WP. Neuropeptide Y is cells: production and effects of corticotropin-releasing hormone. localized together with enkephalins in adrenergic granules of bov- Endocrinology 1997; 138:698–704. ine adrenal medulla. Neuroscience 1988; 25:679–686. 22 Eisenhofer G, Goldstein DS, Stull R, Keiser HR, Sunderland T, Mur- 43 Brimijoin S, Dagerlind A, Rao R, McKinzie S, Hammond P. phy DL, Kopin IJ. Simultaneous liquid-chromatographic determi- Accumulation of enkephalin, proenkephalin mRNA, and neuro- nation of 3,4-dihydroxyphenylglycol, catecholamines, and 3,4- peptide Y in immunologically denervated rat adrenal glands: evi- dihydroxyphenylalanine in plasma, and their responses to inhi- dence for divergent peptide regulation. J Neurochem 1995; 64: bition of monoamine oxidase. Clin Chem 1986; 32: 2030–2033. 1281–1287. 23 Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative 44 Fischer-Colbrie R, Iacangelo A, Eiden LE. Neural and humoral fac- PCR. Genome Res 1996; 6: 986–994. tors separately regulate neuropeptide Y, enkephalin, and chromo- 24 Vale W, Spiess J, Rivier C, Rivier J. Characterization of a 41 residue granin A and B mRNA levels in rat adrenal medulla. Proc Natl ovine hypothalamic peptide that stimulates secretion of corticotro- Acad Sci USA 1988; 85:3240–3244. pin and beta-endorphin. Science 1981; 213: 1394–1397. 45 Stachowiak MK, Hong JS, Viveros OH. Coordinate and differential 25 Koob GF, Heinrichs SC, Menzaghi F, Pich EM, Britton KT. Cortico- regulation of phenylethanolamine N-methyltransferase, tyrosine tropin releasing factor, stress and behavior. Sem in the Neurosci hydroxylase and proenkephalin mRNAs by neural and hormonal 1994; 6: 221–229. mechanisms cultured bovine adrenal medullary cells. Brain Res 26 Sutton RE, Koob GF, Le Moal M, Rivier J, Vale W. Corticotropin 1990; 510:277–288. releasing factor produces behavioural activation in rats. Nature 46 Wan DC, Marley PD, Livett BG. Coordinate and differential regu- 1982; 297:331–333. lation of proenkephalin A and PNMT mRNA expression in cul- 27 Bornstein SR, Ehrhart-Bornstein M, Gu¨se-Behling H, Scherbaum tured bovine adrenal chromaffin cells: responses to cAMP elevation WA. Structure and dynamics of adrenal mitochondria following and phorbol esters. Brain Res Mol Brain Res 1991; 9: 135–142. stimulation with corticotropin releasing hormone. Anat Rec 1992; 47 Eiden LE, Giraud P, Affolter HU, Herbert E, Hotchkiss AJ. Alterna- 234:255–262. tive modes of enkephalin biosynthesis regulation by reserpine and

Molecular Psychiatry Chromaffin cell function in CRHR1-null mice M Yoshida-Hiroi et al 974 cyclic AMP in cultured chromaffin cells. Proc Natl Acad Sci U S ating leptin levels in response to hypersecretion of epinephrine in A 1994; 81:3949–3953. phenylethanolamine-N-methyltransferase (PNMT)-overexpressing 48 Eiden LE, Giraud P, Dave JR, Hotchkiss AJ, Affolter HU. Nicotinic mice. Endocrinology 2000; 141:4239–4246. receptor stimulation activates enkephalin release and biosynthesis 51 Bo¨ttner A, Bornstein SR. Lessons learned from gene targeting and in adrenal chromaffin cells. Nature 1984; 312:661–663. transgenesis for adrenal physiology and disease. Rev Endo Metab 49 Bornstein SR, Tian H, Haidan A, Bo¨ttner A, Hiroi N, Eisenhofer Dis 2001; 2:275–287. G et al. Deletion of tyrosine hydroxylase gene reveals functional 52 Quaife CJ, Hoyle GW, Froelick GJ, Findley SD, Baetge EE, Behringer interdependence of adrenocortical and chromaffin cell system in RR et al. Visualization and ablation of phenylethanolamine N- vivo. Proc Natl Acad Sci U S A 2000; 97: 14742–14747. methyltranferase producing cells in transgenic mice. Transgenic 50 Bo¨ttner A, Haidan A, Eisenhofer G, Kristensen K, Castle AL, Scher- Res 1994; 3: 388–400. baum WA et al. Increased body fat mass and suppression of circul-

Molecular Psychiatry