Growth hormone (GH) autofeedback action in the neonatal rat: involvement of GH-releasing hormone and somatostatin S. G. Cella, V. De Gennaro Colonna, V. Locatelli, V. W. B. Wehrenberg and E. E. M\l=u"\ller

Moiraghi, S. Loche,

Department of Pharmacology, Chemotherapy and Toxicology, University of Milan, Via Vanvitelli 32, 20129 Milan, Italy *First Department of Pediatrics, University of Cagliari, 09100 Cagliari, Italy tDepartment of Health Sciences, University of Wisconsin, Milwaukee, Wisconsin 53021, U.S.A. (Requests for offprints should be addressed to E. E. Müller) received

5 June 1989

ABSTRACT

It is known that in adult rats, GH by itself and by promoting secretion of the somatomedins acts at the level of the hypothalamus to trigger release of somatostatin and decrease output of GH-releasing hormone (GHRH), thereby inhibiting further secretion of GH. To assess whether these mechanisms are already operative in the early postnatal period, we have evaluated the effect of short-term administration of GH in 10-day-old rats. Twice-daily s.c. administration of 25 \g=m\ghuman GH/rat, from days 5 to 9 of life, significantly reduced pituitary content of GH, decreased hypothalamic levels of GHRH mRNA and abolished the in-vivo GH response to a challenge dose of GHRH (20 ng/100 g body weight, s.c.). GHRH (20 ng/100 g

body weight, twice daily, s.c.) given concomitantly with the GH treatment, completely counteracted the inhibitory effect of the latter on pituitary content of GH and restored to normal the in-vivo GH response to the GHRH challenge. These data indicate that impaired secretion of GHRH is involved in the inhibitory effect elicited by GH treatment in infant rats. However, concomitant involvement of hypothalamic somatostatin as a result of GH treatment cannot be ruled out. In fact, pituitaries from rats pretreated with GH responded in the same manner as pituitaries from control rats to the GHRH challenge in vitro. Journal of Endocrinology (1990) 124, 199\p=n-\205

INTRODUCTION

Earlier studies had shown that rats bearing somato¬ trophic tumours have reduced pituitary content of GH and their pituitaries synthesize GH in vitro at a reduced rate (Mac Leod, Fontham & Lehmeyer, 1970; Yamashita, Slanina, Kado & Melmed, 1986). Our present knowledge is limited in that no information is available on the possible feedback action of GH and the underlying mechanism(s) immediately after birth in the rat, a period when the pattern of secretion of GH is rapidly changing. Evidence suggests that the control of GH secretion in the newborn rat is primarily stimulatory, probably due to the presence of low concentrations of immuno¬ reactive somatostatin in the hypothalamus (Walker, Dussault, Alvarado-Urbina & Dupont, 1977) and the limited ability ofsomatostatin to suppress GH secretion both in vitro (Rieutort, 1981; Khorram, De Palatis & McCann, 1983); and in vivo (Rieutort, 1981). Thus, a

Considerable evidence has now been accumulated to indicate that growth hormone (GH) can act to inhibit its own secretion through a complex feedback mechanism operating both on the central nervous system and the pituitary, i.e. GH feeds back at the hypothalamic level to alter secretion of somatostatin and GH-releasing hormone (GHRH). For example, it has been reported previously that somatostatin stores in the rat median eminence are greatly diminished by hypophysectomy (Baker & Yen, 1976), and this depletion is partially prevented by treatment with GH (Hoffman & Baker, 1977). In recent studies, we have shown that in adult male rats, long-term hypophysectomy results in a striking increase in hypothalamic GHRH mRNA, while treatment with GH partially prevents this increase (De Gennaro Colonna, Cattaneo, Cocchi et al. 1988).

study performed in the newborn rat, which also has the advantage of lacking episodic GH release (Cocchi, Gil-Ad, Panerai et al. 1976), offers the possibility of elucidating more clearly the contribution of GHRH in GH negative feedback. To this end, we have investi¬ gated in 5- to 10-day-old rats the effect of a short-term treatment with GH on plasma and pituitary GH con¬ centrations, the pituitary responsiveness to GHRH in vitro and in vivo, the pituitary responsiveness to GHRH in vivo when GHRH is administered concomitantly with GH and the genomic expression of hypothalamic GHRH in the control of GH-treated rats. In a final experiment, pups undergoing short-term treatment with GH were tested for their ability to release GH in response to administration of a somatostatin antiserum. MATERIALS AND METHODS

30 min) and the supernatant was diluted 1 :500 with 001 mol phosphate-buffered saline/1 containing 005 mol EDTA/1 and 1 % (w/v) bovine serum albumin (pH 7-6), and kept frozen until RIA. In-vitro experiments The pituitaries of rat pups treated with hGH or saline were rapidly removed and each pituitary was divided

into four fragments. Eight randomly selected fragments in each group were incubated in plastic vials with 1 ml Krebs-Ringer bicarbonate buffer containing 25 nmol Hepes/1, 10 mmol glucose/1 and 0-1% (w/v) albumin. The vials were incubated for 1 h in a Dubnoffmetabolic shaker under constant gassing with 95% 02 and 5% C02 at 37 °C. At the end of the preincubation period, the medium was replaced with medium containing saline or GHRH (10 nmol/1) and incubated for 4h. Media and tissues were collected after incubation and immediately stored at 20 °C until subsequent RIA. GH released was expressed as per cent of total (released plus intracellular) GH. —

Animals

Infant

(10-day-old) Sprague-Dawley

rats

(Charles

River, Calco, Italy) were used in this study. They were

housed under controlled conditions (22±2°C, 65% humidity and artifical light from 06.00 to 20.00 h). They were obtained with their mothers when 3 days of age; all litters were adjusted to a standard size of 14, and pups were allowed to remain with dams until 1 h before the

experiments.

Experimental procedure

preparation from hypothalamic tissue Four groups of pups (eight animals/group)

RNA

treated

saline were used. Rats were killed brains were rapidly removed and the by decapitation, from each group dissected according to hypothalami a procedure previously described in detail (Cocchi, Gil-Ad, Panerai et al. 1976) and pooled. Tissues were immediately frozen on dry ice and stored at 70 °C until used. Total RNA was isolated from hypothalamic either with GH

or



In-vivo experiments Rat pups were treated in the following manner from 5 to 9 days of age: human GH (hGH; Centro Lombardo Prelievi Terapeutici, Milan, Italy; 25 pg/rat, s.c, at 09.00 and 19.00 h); GHRH (synthetic GHRH(l-29); Sanofi, Paris, France; 20 ng/100 g body weight, s.c. at 09.00 and 19.00 h); concomitant administration of hGH and GHRH as above; control animals were given an isovolumetric amount of normal saline. At the end of treatment and 14 h after the last injection, rats were challenged with saline, GHRH (20 ng/100 g

body weight, s.c), somatostatin antiserum (200 pl/rat i.p.; batch 774; kindly donated by Dr A. Arimura, Tulane University, New Orleans, LA, U.S.A. (Locatelli, Arimura, Torsello et al. 1984)), or coadministration of somatostatin antibody and GHRH. Animals were killed by decapitation 15 min after the

last treatment with saline and GHRH or 60 min after the last treatment with somatostatin antibody. Blood was collected into EDTA-containing tubes, and plasma 20 °C until radio¬ was separated and stored at immunoassay (RIA). Pituitary glands were removed and homogenized in 0-5 ml 001 mol NaHC03/l. The pellets were separated by centrifugation (2500 g for —

by the guanidinium thiocyanate/CsCl method (Glisin, Crkienjakov & Byus, 1974; Chirgwin, Przbyla, MacDonald & Rutter, 1979). For each experimental group, eight hypothalami (about 250 mg frozen tissue) tissue

pooled. Poly (A)+ mRNA was purified by chromatography on oligo-dT cellulose (type 7; Pharmacia, Uppsala, Sweden) (Aviv & Leder, 1972). Analysis of UV absorbance was utilized for RNA quantitation. Starting from 250 mg frozen tissue, 1 ¬ 20 pg total RNA and 5-6 pg poly (A)+ RNA were routinely obtained. were

Slot-blot

analysis

Poly (A)+ RNA samples (2pg each sample) were diluted in 50 pi sterile water. After incubation for 15 min at 65 °C and quick cooling on ice, 50 pi of 10 SCC ( 1 SCC 0-15 mol NaCl/1, 00015 mol Na citrate/1) were added to each sample. The denatured RNAs were spotted in duplicate onto a nitrocellulose sheet (BA85; 0-45 pm; Schleicher & Schuell, Dassel, F.R.G.) (prewetted with 5 SSC) using a slot-blot ap¬ paratus (Minifold II; Scheicher & Schuell). The filters were then baked at 80 °C for 2 h. To determine the =

range of linearity of the autoradiographical signal, known dilutions (8, 1-6, 0-32, 0-064, 0-012 ng) of pBr322 were used and processed similarly to the RNA

and averaged for either the GHcontrol group.

The slot-blot analysis was routinely performed in the presence of control poly (A)+ RNA from a brain

Plasma and pituitary GH contents were determined by a double-antibody RIA (Schalch & Reichlin, 1966). Results were expressed in pg/ml plasma or pg/pituitary in terms of the NIH-2 standard rat GH-RP-2, the potency of which is 20 IU/mg. The sensitivity of the assay was 0-5 ng/1; intra-assay variability was 5-0%. To avoid possible interassay variations, all samples of a given experiment were assayed in a single radioimmunoassay. In our rat GH assay system, cross-reactivity with hGH corresponded to 4-3%.

samples.

(cerebellum) lacking GHRH-producing neurones (unlikely) possibility that the probe utilized could recognize RNAs other than those coding

area

to rule out the

for GHRH. Control of the amount of the RNA blotted was performed by reprobing the slot blots with oligo d(A) d(T) 12-18 (Pharmacia), elongated through terminal transferase with [a32P]dTTP (Amersham International pic, Amersham, Bucks, U.K.). Each experiment was repeated twice. DNA probe and hybridization conditions A plasmid containing the rat GHRH cDNA sequence prGRF2 (Mayo, Cerelli, Rosenfeld & Evans, 1985) (a gift from Dr . E. Mayo, Department of Biochemistry, Northwestern University, Evanston, IL, U.S.A.) was utilized as the probe. The prGRF2 was labelled by Multiprime DNA labelling system (Amersham), with [a32P]dCTP to a specific activity of 2 IO"6 d.p.m./pg cDNA. Nitrocellulose filters were prehybridized in a solution containing 50% formamide (deionized), 5 Denhardt's solution (1 Denhardt's solution 0-02% (w/v) Ficoll, 0-02% bovine serum albumin, 002% (w/v) polyvynylpyrrolidone), 6 SSC, 100 mmol glycine/1 at 42 °C for 4 h. The hybridization was performed in the same solution utilized for the prehybridization ( glycine) and containing the 32P-labelled GHRH plasmid (2 106d.p.m./ml). The hybridization was carried out at 42 °C for 36 h and the filters were washed four times for 30 min in 1 SSC in 0-1 % (w/v) sodium dodecyl sulphate at 50 °C. A signal on X-ray film (Hyper-film HP; Amersham) was detectable after 16 h of exposure in the presence of an intensifying 70 °C. To provide further evidence that screen at data obtained by slot-blot analysis actually reflected changes specifically occurring in GHRH mRNA, a Northern analysis was performed concomitantly with each experiment. Hypothalamic poly (A)+ RNA (5pg) was run on a 1-5% formaldehyde agarose gel, blotted and probed with the specific plasmid. The only band visible after hybridization was of approximately 800 nucleotides (data not shown), a result consistent with published results on the size of hypothalamic GHRH mRNA (Gubler, Monahan, Lomedico et al. 1983; Mayo, Cerelli, Lebo et al. 1985). For quantitative measurement of slot blots, auto=





radiograms

were

subjected

to

scanning densitometry

with an LKB Ultroscan XL Laser Densitometer. The individual densitometric values were normalized to the level of poly (A)+ RNA present in each sample

or

saline-treated

RIA determinations

Statistical

analysis

Plasma and pituitary concentrations of GH in the different experimental groups were compared by Dunnett's /-test preceded by analysis of variance, or by Student's /-test where applicable. A value of

Growth hormone (GH) autofeedback action in the neonatal rat: involvement of GH-releasing hormone and somatostatin.

It is known that in adult rats, GH by itself and by promoting secretion of the somatomedins acts at the level of the hypothalamus to trigger release o...
1MB Sizes 0 Downloads 0 Views