Factor As Determined by Specific Radioimmunoassay Systems
Total Page:16
File Type:pdf, Size:1020Kb
Proc. Natl. Acad. Sci. USA Vol. 82, pp. 2970-2974, May 1985 Medical Sciences Levels of human and rat hypothalamic growth hormone-releasing factor as determined by specific radioimmunoassay systems (growth hormone secretion/growth hormone-releasing factor antiserum/hypothalamic content/ontogeny) TAPAN AUDHYA, MADELINE M. MANZIONE, TOSHICHI NAKANE, NORIO KANIE, JOSEPH PASSARELLI, MARIA Russo, AND CHARLES S. HOLLANDER Division of Endocrinology, Department of Medicine, New York University School of Medicine, 550 First Avenue, New York, NY 10016 Communicated by Saul Krugman, December 17, 1984 ABSTRACT Polyclonal antibodies to synthetic human Many investigators have reported the presence of GH-re- pancreatic growth hormone-releasing factor [hpGRF(l- leasing activity in human carcinoid tumors (13-15) and pan- 44)NH2] and rat hypothalamic growth hormone-releasing fac- creatic islet cell tumors removed from patients with acro- tor [rhGRF(1-43)OHJ were produced in rabbits by injecting megaly (15-17). Recently, human pancreatic GRF (hpGRF) these weak immunogens, coupled to thyroglobulin and emulsi- was isolated and structurally characterized from two pancre- fied with complete Freund's adjuvant in the presence of acti- atic islet cell tumors in patients with acromegaly (9, 18, 19). vated charcoal, directly into the spleen. A subsequent booster In one tumor, only a 40-amino acid peptide was isolated (18, in-jection by the conventional intramuscular route resulted in 19), while the other tumor was found to contain three pep- high-titer antibodies, which at a 1:20,000 dilution were used to tides consisting of 37, 40, and 44 amino acids (9), of which develop highly sensitive and specific radioinmunoassays for the 40-amino acid peptide was the major component. Recent these peptides. By using antibodies with an apparent K. of 3.3 evidence indicates that human hypothalamic GRF is identi- x 10-12 (human) and 7.7 x 10-11 (rat), the sensitivity of these cal to the 44-amino acid COOH-terminal amidated hpGkF assays in both human and rat was found to be <1 fmol. The (refs. 12, 20; *). Hypothalamic GRF has been characterized antibody to hpGRF(1-44)NH2 is directed against the COOH- in many species-porcine (21), bovine (22), and ovine (12)- terminal region of the molecule, as shown by its crossreactivity as 44-amino acid COOH-terminal amidated peptides with with various hpGRF analogues: 140% with hpGRF(30- <5% amino acid substitution from hpGRF(1-44)NH2.* 44)NH2; 1%-2% with hpGRF(1-37)OH, hpGRF(140)OH, Spiess et al. (23) have reported that rat hypothalamic GRF and hpGRF(1-40)NH2; and none with hpGRF(1-29)NH2. Se- (rhGRF) is a 43-amino acid NH2-terminal histidine, COOH- rial dilutions of human and rat hypothalamic extracts demon- terminal free acid polypeptide, with a 67% homology with strated parallelism with the corresponding species-specific hpGRF(1-44)NH2. These GH-releasing peptides have a standard and 12,5-labeled tracer. There was no crossreactivity close structural homology with the glucagon-secretin family, with other neuropeptides, gastrointestinal peptides, or hypo- especially PHI-27 (18, 23-25). thalamic extracts of other species. The hypothalamic content Both native and synthetic GRF peptides have been shown in fmol/mg (wet weight) of tissue was 3.6 ± 0.2 for the human to be specific and potent stimulators of GH secretion from and 11.1 ± 5.5 for the rat. Age-related changes in hypotha- the anterior pituitary in vitro and in vivo (9, 18, 26, 27). In lamic GRF content were present in rats, with a gradual in- addition, the NH2-terminal 29 amino acids of hpGRF main- crease from 2 to 16 weeks and a correlation between increasing tain full intrinsic activity and potency in vitro (18), and re- body weight and GRF content. These radioimmunoassays winl cently amino acids 3-21 of human hypothalamic GRF have serve as important tools for understanding the regulation of been shown to contain complete intrinsic GRF activity.* growth hormone secretion in both human and rat. To evaluate the physiological role of these GH-releasing peptides, we raised antibodies to synthetic hpGRF(1- Since 1959 (1), it has been known that the hypothalamus is 44)NH2 and rhGRF(1-43)OH. In this report, we describe the involved in the control of growth hormone (GH) secretion. production of these high-affinity polyclonal antibodies and Experimental and clinical evidence has since indicated that their use in radioimmunoassay systems for the detection of the hypothalamus performs a dual function in the regulation species-specific GRF-like immunoreactivity in extracts of of GH secretion from the anterior pituitary (2-4). Inhibition hypothalamic tissue as well as delineation of age-related of GH release is accomplished by an inhibitory peptide, so- changes in hypothalamic GRF content in the rat. matostatin (SRIF), which in 1972 was isolated from ovine hypothalami, purified, and characterized as a tetradecapep- METHODS tide (5). In contrast, as suggested by physiological evidence (2-4) and GH-releasing activity in hypothalamic extracts (6- Preparation of Immunogen. Thyroglobulin was conjugated 8), stimulation of GH release is accomplished by a stimula- to two synthetic GRF peptides by making two solutions con- tory peptide, GH-releasing factor (GRF). Until 1982, howev- taining 1 mg of bovine thyroglobulin (Sigma), 1 mg of syn- er, the isolation and characterization of GRF remained elu- thetic peptide [hpGRF(1-4)NH2 or rhGRF(1-43)OH] dis- sive. The difficulty in its characterization had been due to solved in 1 ml of 10 mM phosphate buffer (pH 7.4), and 8 Al the minute quantities of the peptide in hypothalamic tissue, of 25% (vol/vol) glutaraldehyde solution (Sigma). Each solu- the large quantities of SRIF that interfere in the bioassay, and the fact that the methodology used for peptide sequenc- Abbreviations: GH, growth hormone; hpGRF, human pancreatic ing and purification has only recently become available (9- growth hormone-releasing factor; rhGRF, rat hypothalamic growth 12). hormone-releasing factor; SRIF, somatostatin; CRF, corticotropin- releasing factor. *Guillemin, R., Ling, N., Esch, F., Bohlen, P., Brazeau, P., Weh- The publication costs of this article were defrayed in part by page charge renberg, W., Baird, A., Bloch, B., Mougin, C. & Zeytin, F., Sev- payment. This article must therefore be hereby marked "advertisement" enth International Congress on Endocrinology, July 1-7, 1984, in accordance with 18 U.S.C. §1734 solely to indicate this fact. Quebec City, PQ, Canada, symp. 165. 2970 Downloaded by guest on September 23, 2021 Medical Sciences: Audhya et aL Proc. Natl. Acad. Sci. USA 82 (1985) 2971 tion was stirred on a rotary mixer for 1 hr at room tempera- was added to the appropriate antibody tubes after 48 hr of ture, then dialyzed in spectapor 3 membrane tubing (Spec- cold incubation; radiolabeled rhGRF(1-43)OH (2500 cpm trum Medical Industries, Los Angeles) against a 1-liter 10 per 100 /l) was added after 24 hr of cold incubation. After mM phosphate buffer (pH 7.4) for 2 hr at room temperature incubation at 40C for 5 days for hpGRF(1-44)NH2 and for 3 with constant stirring. The dialyzed complexes were emulsi- days for rhGRF(1-43)OH, the antigen-antibody complex fied with complete Freund's adjuvant (GIBCO) and normal was separated from the free antigen by dextran-coated char- saline (1:1:1) in the presence of 10 mg of active charcoal coal as described. The bound-to-free (B/F) values were cal- (Fisher), using two glass syringes connected by a three-way culated and plotted against the corresponding antibody dilu- stopcock until the emulsion was firm. tion. The affinity constant (Ka) for each titration curve was Immunization. Female New Zealand White rabbits (Hazel- determined according to the method of Frazier and Rodbard ton Dutchland, Denver, PA) were immunized with one of the (30). complexes by the following procedure: The rabbits were Tissue Extraction. Rat hypothalami were obtained from anesthetized with sodium pentobarbitol and atropine sulfate. adult male Sprague-Dawley rats (200-250 g) (Taconic The skin overlying the region of the spleen was shaved, ster- Farms, Germantown, NY) immediately after decapitation ilized with betadine solution, and anesthetized with 2% lido- and were kept at -20'C. Mouse and guinea pig hypothalami caine containing epinephrine (1:100,000). A left lateral inci- were collected in the same manner from 56- to 70-day-old sion was made and the spleen was exposed and injected with male BALB/c mice (Charles River Breeding Laboratories) a total of 450 ,1 of the freshly conjugated peptide/thyroglo- and young adult male Hartley white guinea pigs (400-500 g) bulin/Freund's adjuvant emulsion (150 jxg of peptide). The (Camm Research Institute, Black Oak Ridge, NJ). Bovine incision was then closed in a layered fashion. Two weeks and ovine hypothalami were obtained as freshly frozen tis- after the initial immunization, the rabbits were given intra- sue (Pel-Freeze) and human hypothalami were obtained in muscular booster injections of 1 ml of the complex (100 gg of the course of autopsy examination of healthy male individ- peptide) in the presence of incomplete Freund's adjuvant. At uals 6-12 hr after accidental death. Tissues from all speci- 1- to 2-week intervals, blood was collected from the ear vein mens were homogenized with distilled water at a concentra- and the serum was examined for the presence of antibodies. tion of 100 mg of tissue per ml of water. The homogenized When a consistent decline in antibody titer was observed, suspension was then heat-treated in a boiling water bath for intramuscular booster injections were repeated. 10 min, chilled at 40C, and centrifuged at 12,400 x g at 40C Radioimmunoassays.