Corticotropin releasing factor binding protein (CRF-BP) is expressed in neuronal and astrocytic cells

BRAIN RESEARCH ELSEVIER

Brain Research 698 (1995) 259-264

Short communication

Corticotropin releasing factor binding protein (CRF-BP) is expressed in neuronal and astrocytic cells Dominic P. Behan *, Dominique Maciejewski, Derek Chalmers, Errol B. De Souza Neurocrine Biosciences Inc., 3050 Science Park Rd., San Diego, CA 92121, USA Accepted 1 August 1995

Abstract

Corticotropin releasing factor (CRF) binding protein (CRF-BP) was measured in media and cell lysates of primary rat astrocytes, microglia and neurons with the use of a ligand immunoradiometric assay (LIRMA). A low basal level of CRF-BP was detected in the media and cell lysates from primary neuronal and astrocyte cells after 48 h in culture. No basal expression of CRF-BP was detected in cell lysates or media from primary microglial cultures. The CRF-BP expressed in cultured astrocytes and neurons had the same pharmacological characteristics as the human recombinant molecule. After forskolin, IBMX or forskolin/IBMX treatment, a robust increase in secreted CRF-BP levels in the media from astrocytes and neurons, but not microglia, was observed. An increase in CRF-BP-Iike immunoreativity in cell lysates was also observed after IBMX/forskolin treatment. In situ hybridization analysis revealed that CRF-BP mRNA was increased in primary cultured astrocytes after IBMX/forskolin stimulation suggesting that regulation was at the level of gene transcription. 'Axon sparing' lesions produced with 0.12 M quinolinic acid in PBS injected intracerebrally (unilaterally into dorsal hippocampus) resulted in loss of CRF-BP expression in neuronsl These data provide evidence for the differential localization and regulation of CRF-BP in different cell types in brain and suggest that CRF-BP expression may be locally increased in disease states associated with astrocytosis and gliosis. Keywords: Corticotropin releasing factor binding protein; Expression; Neuronal cell; Astrocytic cell

Corticotropin releasing factor (CRF), a 41-residue peptide has many different roles all of which share a common theme of adapting the b o d y ' s behavioral, endocrine, autonomic and immune responses to stress. In addition to its primary endocrine effects to stimulate proopiomelanocortin derived peptides from the pituitary [27] CRF has been shown to be anxiogenic in animals [14], to improve arousal and learning [15,8], to modulate food intake [24] and thermogenesis [25], to alter blood pressure [9,16] and to have both anti- [13] and pro-inflammatory [6,5] effects depending on its site of action. In addition to actions at membrane receptors, CRF binds, with high affinity to a 37 kDa protein which has been purified from human plasma [1] and subsequently cloned from a human liver Agtll c D N A Library [22]. CRF-BP inhibits the adrenocorticotrophic hormone (ACTH) releasing properties of CRF in vitro and protects the maternal pituitary gland from elevated plasma CRF

* Corresponding author. Fax: (1) (619) 658-7602. 0006-8993/95/$0%50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 1 0 1 4 - 0

levels found during the third trimester of human pregnancy [19,17,26,18]. CRF-BP is also broadly distributed in rat brain and pituitary as demonstrated by in situ hybridization and immunohistochemistry [23]. In the pituitary gland, CRF-BP was localized to pituitary corticotropes, where it may play a paracrine role in the regulation of the stress response. In brain, CRF-BP is expressed predominantly in the cerebral cortex, including all major archi, paleo, and neocortical fields. CRF-BP expression was also found in several raphe nuclei, a number cell groups in the brainstem reticular core, the amygdala and bed nucleus of the stria terminalis and sensory relays associated with the olfactory, auditory, trigeminal and vestibular systems. No CRF-BP expression was detected in the paraventricular nucleus of the hypothalamus and in this region, expression was limited only to the premammillary and dorsomedial nuclei. While the pattern of distribution in brain appears to reflect a primary neuronal localization of the protein, the precise cellular (i.e. neuronal a n d / o r glial) localization remains to be demonstrated. The purpose of the current study was to

260

D.P. Behan et al. / Brain Research 698 (1995) 259-264

establish if CRF-BP is expressed by neuronal and/or glial cell types such as astrocytes and microglia. The astrocyte cultures were prepared as previously described [20]. In brief, the cerebral hemispheres were dissected out from 1-2-day-old newborn rat pups and stored in Tyrode's solution at room temperature. The meninges were carefully removed, and the tissue mechanically dissociated by successive passages through needles of decreasing gauges [18,21,25]. The resulting cell suspension was then diluted in DMEM containing 10% fetal calf serum (FCS), plated in 72 cm 2 flasks at a density of 105 cells/cm 2 and incubated at 37°C in an atmosphere of 95% air/5% CO~ and 100% humidity. The medium was changed every 3 - 4 days until the cells reached confluence (10-12 days). To obtain purified astrocytes, confluent mixed glial cells were shaken overnight at 200 rpm, 37°C, to dislodge the overlaying microglia and oligodendrocytes. They were subsequently treated with 10 mM leucine-methylesther (LME) in serum-free DMEM and shaken at 200 rpm, 37°C for 6 h. The medium was replaced overnight with DMEM + 10% FCS and the same treatment with LME was repeated the next day. The cells were then detached by incubating them for 20 rain in 0.05% trypsin and replated in 24-well trays at a density of 105 cells/cm 2. The cells were at least 95% pure astrocytes as verified by GFAP staining. The astrocytes were assayed 3 - 5 days after replating. In some cases, the confluent mixed glial cells were left without changing the medium for at least seven days. They were then shaken overnight at 200 rpm, 37°C, the supernatant was collected for microglial cultures and fresh medium was added to the remaining astrocytes. Again, they were left for 7 days without medium change, then shaken overnight, treated with LME and replated as for purified astrocytes. They were subjected to various treatments 2 days after replating. These old astrocytes were maintained in culture for a total of 30-35 days before assay. Type 2 astrocyte enriched cultures were obtained following the method as previously described [7]. Briefly mixed glial cell cultures were plated in DMEM + 20% FCS for three days, then shaken for 1 h at 100 rpm. Medium was replaced with DMEM + 10% FCS and the cells were allowed to grow for 4 more days. They were then shaken for 16 h at 100 rpm, the supernatant was filtered through a 40 p~m filter and spun for 5 min at 90 × g. The resulting pellet was dispersed in DMEM containing 10% FCS and 10 n g / m l fibroblast growth factor and plated in a 24-well dish. The type 2 astrocytes were highly positive for GFAP staining and displayed a typical stellate morphology. Microglial cultures were established as previously described [10] with a modification suggested by Hao et al. [12]. In brief, after starving the mixed glial cultures for 7 - 1 0 days, the cells were shaken overnight at 200 rpm, 37°C. The supernatant was collected filtered through a 70

p~m sieve and spun down for 10 min at 300 × g. The cell pellet was resuspended in DMEM with 10% FCS, subjected to a cell count and replated at a density of 105 cells/cm 2. Two to three hours later, the cells were manually shaken and the medium changed with DMEM containing 10% FCS and 100 ~1 of Macrophage stimulating colony factor (M-CSF) and Granulocyte-Macrophage colony stimulating factor (GM-CSF). These cells were positive for MAC-l, incorporated Dil-ac-LDL and were negative for GFAP (astrocytes) and Galactose cerebroside (oligodendrocytes). Dispersed cortical cultures were derived from 15-dayold fetal Sprague-Dawley rats. Dissected cortical tissue was collected in Tyrodes' solution, then dissociated in DMEM containing 0.6% glucose, 0.2% L-glutamine, 100 U / m l penicillin, 100 /~g/ml streptomycin and 10% FCS by successive passage through 18-, 21- and 25-gauge needles. The cell suspension was then filtered through a 70 /~m nylon sieve, and plated on poly-L-lysine coated 24-well trays at a density of 250,000 cells/cm 2. The next day, half of the medium was replaced by medium containing cytosine arabinoside (5 /xM). Half of the medium was then replaced every 2 - 3 days. The cultures were usually treated after 6 days in vitro. All cell types were stimulated with 25 /~M forskolin and/or 250 /~M IBMX after six days in culture. After stimulation, the media was collected and the cells were scraped from the wells in 100 pA of PBS and solubilized with the addition of 25 /~1 of 1% NP-40. CRF-BP was measured in media and cell lysates by ligand immunoradiometric assay (LIRMA). Briefly, dilutions of each sample or purified CRF-BP (50 ~1) were added to 200/xl of assay buffer (50 mM sodium phosphate buffer containing 0.02% NP-40) in glass silicate tubes. [125I]Tyr°-rat/human CRF ([125I]CRF; DuPont NEN: specific activity 2200 /xCi/mmol) (45 p M / 5 0 ~1 in assay buffer) was then added to each tube followed by incubation for 60 min at room temperature to allow binding to occur. CRF-BP antibody 5144 (50 /M; diluted 1:1000 with assay buffer) was then added to each tube and the reaction was subsequently incubated overnight at room temperature. Competitive binding curves were obtained by performing the assay in the same way as the LIRMA but in the presence of increasing concentrations of unlabeled competitor. In all cases the total volume of all tubes was made to 300 /zl with assay buffer. Bound complexes were precipitated by the addition of 200/xl of precipitated sheep anti-rabbit (SAR) second antibody (a mixture containing sheep anti-rabbit IgG (SAR) 1:20, 1% normal rabbit serum (NRS), 4% PEG, 50 mM sodium phosphate, 0.1% sodium azide), followed by incubation for 60 min at room temperature. The antibody-bound ~25I-CRF precipitate was then separated by centrifugation (3000 × g] at 4°C for 20 min. Riboprobes were produced using either T3 or T7 transcription systems in a standard labeling reaction mixture consisting of: 1 /.zg linearized plasmid, 5 × transcription

D.P. Behan et al./Brain Research 698 (1995) 259-264

buffer, 125 mCi [35S]UTP or [33p]UTP, 150 /zM NTP's, 12.5 mM dithiothreitol, 20 U RNAase inhibitor and 6 U of the appropriate polymerase. The full-length hCRF-BP cDNA [22] cloned into the EcoRI site of pBluescript SK was used. The insert was oriented such that the T3 direction was in the sense direction. The reaction was incubated at 37°C for 90 min, labeled probe being separated from free nucleotides over Sephadex G-50 spin columns. For in situ hybridization and immunocytochemistry, all the different cell types were plated on Tekto 8-well glass slides, grown as their counterpart and fixed in 4% paraformaldehyde before appropriate staining. Dissected tissue was frozen in isopentane cooled to - 4 2 ° C and subsequently stored at - 8 0 ° C prior to sectioning on a cryostat. Slide-mounted tissue sections were then stored at - 8 0 ° C . Sections were removed from storage and placed directly into 4% buffered paraformaldehyde at room temperature. After 60 min, slides were rinsed in isotonic phosphate buffered saline (10 min) and treated with proteinase K (1 /xg/ml in 100 mM Tris-HC1, pH 8.0) for 10 min at 37°C. Subsequently, sections underwent successive washes in water (1 min), 0.1 M triethanolamine (pH 8.0, plus 0.25% acetic anhydride) for 10 min and 2 × SSC (0.3 mM NaCI, 0.03 mM sodium citrate, pH 7.2) for 5 min. Sections were then dehydrated through graded alcohols and air dried. Post-fixed sections were hybridized with 1.0 × 10 6 dpm [35S]UTP-labeled riboprobes in hybridization buffer containing 75% formamide, 10% dextran sulphate, 3 × SSC, 50 mM sodium phosphate buffer (pH 7.4), 1 × Denhardt's solution, 0.1 m g / m l yeast tRNA and 10 mM dithiothreitol in a total volume of 30 /zl. The diluted probe was applied to sections on a glass coverslip and hybridized overnight at 55°C in a humid environment. Post-hybridization, sections were washed in 2 × SSC for 5 min and then treated with RNAase A (201) /xg/ml in 10 mM Tris-HC1, pH 8.0, containing 0.5 M NaCI) for 60 min at 37°C. Subsequently, sections were washed in 2 × SSC for 5 min, 1 × SSC for 5 rain, 0.1 × SSC for 60 min at 70°C, 0.5 × SSC at room temperature for 5 min and then dehydrated in graded alcohols and air dried. For signal detection, sections were placed on Kodak BioMax X-ray film and exposed for the required length of time or dipped in photographic emulsion (Amersham LM-1) for high resolution analysis. Auto ra-

261

Table 1 IC50 values for various peptides to the CRF-BP Peptide

Astrocytes (nM)

Neuronal cultures (nM)

Recombinant hCRF-BP (nM)

r / h CRF 0.2 o~-helical 0.32 oCRF(9-41 )

[).23 [I.54

I). l t? I).2

r/h CRF(6-33)

0.82

2.8

~).gt?

oCRF

> 500

> 500

471

diograms were analyzed using automated image analysis (DAGE c a m e r a / M a c II) while dipped sections were examined using a Zeiss Axioscope. Neuron-specific 'axon sparing' lesions were produced in three rats (250-300 g) with 0.12 M quinolinic acid in PBS injected unilaterally into the dorsal hippocampus in a final volume of 1 /.tl. Injection of PBS in the contralateral hippocampus served as control. Animals were sacrificed 5 days post-injection. A low basal level of CRF-BP was detected by LIRMA in conditioned media from primary astrocytes and neurons (Fig. 1). After treatment of the cells with IBMX (250 /xM)/forskolin (25 /,zM) a large increase in CRF-BP secretion into the media was detected in astrocytes and mixed neuronal cultures but not microglia (Fig. 1). We consistently observed that far higher levels of CRF-BP were produced in old astrocytes which had been kept in culture for a total of 25-30 days and had been starved of media changes for periods of 7 days. The media starvation enhances the proliferation of microglia such that the overall proportion of contaminating microglia in the old astrocyte cultures was much higher than in the young cultures. We therefore believe that astrocytes need to be in an activated state to produce large amounts of CRF-BP. Despite the lack of CRF-BP production from microglia it remains a possibility that factors produced from microglia are required to stimulate CRF-BP production from the old astrocytes such that gliosis may play an indirect role in stimulating astrocytic CRF-BP expression in vivo. Total CRF-BP levels secreted into the media rose to 45 ng in astrocyte cultures and 37 ng in mixed neuronal

50 40-

6 --0- Cells

] •4 3

~30~'202

100

, M Control

, M + IfF

A Control

I

0

I

A + I/F

N Control

N I/F

Fig. 1. Expression of CRF-BP in type I astrocytes (A), microglia (M) and neurons IN) before and after I B M X / f o r s k o l i n ( l / F ) stimulation. ('ontrul is media alone.

262

D.P. Behan et al. / Brain Research 698 (1995) 259-264

50,

,,,,,0--

Cells

[

,1.5

.~

. ~ 40-

.

r,,9

~.~. e~

•

-0.5

i

~

~ tO-

| '1"2 Control

r T2 + 1

!

r T2 + F

|

T T 2 + [/F

T1 C o n t r o l

T I + I/F

Treatment

Fig. 2. Comparison of expression of CRF-BP in type 1 (TI) and type 2 (T2) astrocytes after stimulation with forskolin (F), IBMX (1) or IBMX/fbrskolin (I/F). Control is media alone.

cultures (Fig. 1). There was little evidence for any increase in CRF-BP levels in the cell lysates from astrocytes after IBMX/forskolin stimulation suggesting that the protein is rapidly secreted into the media (Fig. 1). We consistently found the ratio of total CRF-BP expressed in media: total CRF-BP expressed in cells to be less in mixed neuronal cultures than in astrocytic cultures suggesting that there is more cell/membrane associated CRF-BP in neurons than in astrocytes (Fig. 1). This is consistent with previous observations that the brain CRF-BP is predominently membrane associated by virtue of its differential localization to membrane pellets after centrifugation of rat, sheep [3] and human [4] brain homogenates. However, most of the total CRF-BP activity was detected in the media from both cultured astrocytes and neurons suggesting that a secreted form of the protein is made during cell activation with cAMP. This is in conflict with in vivo data demonstrating an absence of CRF-BP in human and rat CSF (unpublished observations) but it is possible that CRF-BP may be secreted into the CSF under circumstances where astrocytes and the CRF-BP neurons are activated such as when CRF is increased in the brain following astrocytosis or stress. Also, it was evident in separate experiments that

cells which stained positive for CRF-BP with an anti-CRFBP antibody could also be stained for neurofilament demonstrating localization of the CRF-BP to neurons (data not shown). However, we were unable to obtain positive staining for CRF-BP in astrocytes using immunocytochemistry. The rapid release of the binding protein from astrocytes as measured by immunoassay is likely to be responsible for the lack of cell staining with the anti-CRF-BP antibody. Furthermore, the positive staining of neurons with the anti-CRF-BP antibody agrees with the increased CRF-BP levels detected in cell lysates from neurons (Fig. 1) and further supports the possibility of a membrane associated form of the protein in this cell type. The CRF-BP activity found after stimulation of astrocytes with IBMX/forskolin originated purely from type 1 astrocytes as stimulation of type 2 astrocytes gave no evidence for increased CRF-BP expression (Fig. 2). Of note, type 1 astrocytes are the only astrocytes known to exist in vivo [21]. Controversy still exists as to whether type 2 astrocytes exist in vivo or if they are produced by an artifact of cell culture preparation [21]. CRF-BP in media from astrocytes and mixed neuronal cultures showed an identical rank order binding profile to

Fig. 3. In situ hybidization of CRF-BP mRNA in cultured astrocytes before (A) and after (B) IBMX/forskolin stimulation.

D.P. Behan et al. / Brain Research 698 (1995) 259-264

CONTROL

:~ B

263

! EStONED

D(3

.:

Fig. 4. In situ hybridization of CRF-BP mRNA in the hippocampus following (A) PBS injection and (B) quinolinic acid injection in the contralateral hemisphere to (A). Note the lack of CRF-BP mRNA expression in both the CAI and dentate gyrus after quinolinic acid lesion (B). * Indicates tissue artifact.

the human recombinant molecule ( r / h CRF = c~-helical o C R F ( 9 - 4 1 ) > r / h C R F ( 6 - 3 3 ) >> o C R F (Table 1). These data suggest that the CRF-BP produced from astrocytes and neurons originates from the same gene as the currently cloned CRF-BP [22]. After probing cultured astrocytes and neurons with a c R N A probe directed to the full length CRF-BP, a large increase in CRF-BP m R N A expression was detected in astrocytes and neurons which had been treated with I B M X / f o r s k o l i n suggesting that regulation was occurring at the level of transcription (Fig. 3). There was little evidence for any CRF-BP m R N A expression in the absence of the I B M X / f o r s k o l i n stimulation. It was evident in separate experiments that cells which expressed CRF-BP m R N A could be localized for neurofilament further demonstrating localization of the CRF-BP to neurons (data not shown). However, it remains possible that I B M X / f o r skolin a n d / o r c A M P treatment may also act to increase the stability of CRF-BP mRNA. Nevertheless these data clearly indicate that astrocytes are capable of expressing CRF-BP m R N A in vitro. The ability of forskolin to stimulate CRF-BP expression in neurons and astrocytes may not be a direct c A M P response element binding protein (CREB)-mediated event [11] on the CRF-BP promoter as there was no c A M P response element found in the human promoter sequence [2]. It is possible that the rat promoter differs in this respect or that the c A M P effect is mediated through a pathway which does not utilize CREB. Within the dorsal hippocampus, CRF-BP m R N A was expressed in scattered cells in C A fields and within the dentate gyrus (Fig. 3). Within the CA1 subfield CRF-BPexpressing cells were evident in pyramidal and non-pyramidal cell layers. Post-quinolinic acid injection, CRF-BP m R N A was undetectable in both CA1 and dentate gyrus (Fig. 4) indicating that CRF-BP is normally localized to neurons in this region. Thus, within the brain CRF-BP is likely to be expressed in both neurons and astrocytes. It is possible that the astrocytic source of CRF-BP plays some role in regulating synaptic CRF originating from neurons when astrocytes are activated. Thus, CRF-BP may be elevated in brain in

inflammatory disease states associated with astrocytosis and gliosis, such as Alzheimer's disease. In light of the local proinflammatory effects of CRF in tissues and joints [6,5], it is possible that increased CRF-BP expression occurs locally at specific lesion sites to counteract increased CRF production. The specific roles of these "pools" of CRF-BP in relation to the functionality of CRF-related pathways remain to be determined.

Acknowledgements We would like to thank Dr. Stephen Heinrichs for help with the quinolinic acid lesion and Dr. Wylie Vale for providing the full length human CRF-BP c D N A probe.

References [1] Behan, D.P., Linton, E.A. and Lowry, P.J., Isolation of the human plasma corticotrophin-releasing factor-binding protein, J. Endocrinol., 122 (1989) 23-31. [2] Behan, D.P., Potter, E., Lewis, K.A., Jenkins, N.A., Copeland, N., Lowry, P.J. and Vale, W.W., Cloning and structure of the human corticotrophin releasing factor-binding protein gcne (CRHBP), Genomics, 16 (1993) 63-68. [3] Behan, D.P., Potter, E., Sutton, S., Fischer, W., Lowry, P.J. and Vale, W.W., Corticotropin-releasing factor-binding protein: a putative peripheral and central modulator of the CRF family of neuropeptides, Ann. NYAcad. Sci., 697 (1993) 1 8. [4] Behan, D.P., Troncoso, J.C., Ling, N. and De Souza, E.B., Corticotropin-releasing factor-binding protein in human brain: identification and characterization in Alzheimer's disease. In 24th Annual Meeting of the Society for Neuroscience, Miami. FL. 1994, Abstr. 552.10. [5] Crofford, L.J., Sano, H., Karalis, K., Friedman, T.C., Epps, H.R., Remmers, E.F., Mathern, P., Chrousos. G.P. and Wilder, R.L., Corticotropin-releasing hormone in synovial fluids and tissues of patients with rheumathoid arthritis and osteoarthritis, J. Immunol., 151 (1993) 1587-1596. [6] Crofford, L.J., Sano, H., Karalis, K., Webster, E.L., Goldmuntz, E.A., Chrousos, G.P. and Wilder, R.L., Local secretion of corticotropin-releasing hormone in the joints of Lewis rats with inflammatory arthritis, J. C lin. lnL'est. , 90 (1992) 2555-2564. [7] Da Cunha, A. and Vitkovic, L., Simultaneous preparation of four cell type enriched glial cultures from rat cerebral cortex, J. Tiss. Cult, Methods, 13 (1991) 31-38.

264

D.P. Behan et al. / Brain Research 698 (1995) 259-264

[8] Diamant, M. and De Wied, D., Structure-related effects of CRF and CRF-derived peptides:dissociation of behavioral, endocrine and autonomic activity, Neuroendocinology, 57 (1993) 1071-1081. [9] Fisher, L.A., Rivier, J., Rivier, C., Speiss, J., Vale, W.W. and Brown, M.R., Corticotropin-releasing factor (CRF): Central effects on mean arterial pressure and heart rate in rats, Endocrinology, 110 (1982) 2222-2224. [10] Giulian, D. and Baker, T., Characterization of ameboid microglia isolated from developing mammalian brain, J. Neurosci., 6 (1986) 2163-2178. [11] Gonzalez, G., Yamamoto, K., Fischer, W.H., Karr, D., Menzel, P., Biggs, W., Vale, W.W. and Montminy, M.R., A cluster of phosphorylation sites on the cyclic AMP-regulated nuclear factor CREB predicted by its sequence, Nature, 337 (1989) 749-752. [12] Hao, C., Richardson, A. and Fedoroff, F., Macrophage-like cells originate from neuroepithelium in culture: characterization and properties of the macrophage-like cells, Int. J. DeLl., 9 (1990) 1-14. [13] Kiang, J.G. and Wei, E.T., Corticotropin-releasing factor inhibits thermal injury, J. Pharmacol. Exp. Ther., 243 (1987) 517-520. [14] Koob, G.F., Stress, corticotropin-releasing factor and behavior, Perspect. BehaL,. Med., 2 (1985) 39-52. [15] Koob, G.F. and Bloom, F.E., Corticotropin-releasing factor and behavior, Fed. Proc., 44 (1985) 259-263. [16] Lenz, H.J., Fisher, L.A., Vale, W.W. and Brown, M.R., Corticotropin-releasing factor, sauvagine and urotensin I: effects on blood flow, Am. J. Physiol., 249 (1985) R85-R90. [17] Linton, E.A., Behan, D.P., Saphier, P.W. and Ii)wry, P.J., Corticotropin-releasing hormone binding protein: reduction in the ACYH-releasing activity of placental but not hypothalamic CRH, J. Clin. Endocrinol. Metab., 70 (1989) 1574-1580. [18] Linton, E.A., Perkins, A.V., Woods, R.J., Eben, F., Wolfe, C.D.A., Behan, D.P., Potter, E., Vale, W.W. and Lowry, P.J., Corticotropinreleasing hormone-binding protein (CRH-BP): plasma levels de-

[19]

[20]

[21] [22]

[23]

[24]

[25] [26]

[27]

crease during the third trimester of normal human pregnancy, .I. Clin. Endocrinol. Metab., 76 (1993) 260-262. Linton, E.A., Wolfe, C.D.A., Behan, D.P. and Lowry, P.J., A specific carrier substance for human corticotropin-releasing factor in late gestational maternal plasma which could mask the ACTH-releasing activity, Clin. Endocrinol., 28 (1988) 315-324. McCarthy, K. and De Vellis, J., Preparation of separate astroglial and oligodendroglial cell cultures from rat cerebral tissue, J. Cell Biol., 85 (1980) 890-902. Nobel, N., Points of controversy in the O-2A lineage: clocks and type-2 astrocytes, Glia, 4 (1991) 157-164. Potter, E., Behan, D.P., Fischer, W.H., Linton, E.A., Lowry, P.J. and Vale, W.W., Cloning and characterization of the cDNAs for human and rat corticotropin-releasing factor-binding proteins, Nature, 349 (1991) 423-426. Potter, E., Behan, D.P., Linton, E.A., Lowry, P.J., Sawchenko, P.E. and Vale, W.W., The central distribution of a corticotropin-releasing factor (CRF)-binding protein predicts multiple sites and modes of interaction with CRF, Proc. Natl. Acad. Sci. USA, 89 (1992) 4192-4196. Richard, D., Involvement of corticotropin-releasing factor in the control of food intake and energy expenditure, Ann. NY Acad. Sci., 697 (1993) 155-172. Rothwell, N.J., Central effects of CRF on metabolism and energy balance, Neurosci. Behat,. Ret:., 14 (1990) 263-271. Sada, T., Iwashita, M., Tozawa, F., Ushiyama, Y., Tomori, N., Sumitomo, T., Nakagami, Y., Demura, H. and Shizume, K., Characterization of CRH binding protein in human plasma by chemical cross-linking and its binding during pregnancy, J. Clin. Endocrinol. Metub., 67 (1988) 1278-1283. Vale, W., Spiess, J., Rivier, C. and Rivier, J., Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and /3-endorphin, Science, 213 (1981) 1394-1397.