Characterization of growth hormone-releasing hormone (GHRH) binding to cloned porcine GHRH receptor

Peptides, Vol. 16, No. 8, pp. 1469-1473, 1995 Copyright 0 1995 Elsevier Science Inc. Printed in the USA. All rights reserved 0196.9781/95 $9.50 + .OO

Pergamon

01%-9781(95)02026-S

Characterization of Growth Hormone-Releasing Hormone (GHRH) Binding to Cloned Porcine GHRH Receptor HAZEM

A. HASSAN,

HANSEN M. HSIUNG, XING-YUE ZHANG, DAVID L. SMILEY AND MARK L. HE&IAN’

DENNIS

P. SMITH,

Division of Endocrinology, Eli Lilly csiCompany, Indianapolis, IN 46285 Received HASSAN,

H. A., 1-I. M. HSIUNG,

X.-Y. ZHANG,

5 May 1995

D. P. SMITH,

D. L. SMILEY

AND M. L. HEIMAN. Characrerizarion of 16(8) 1469-1473, 1995.To study structure-activity relationships of growth hormone-releasing hormone (GHRH), a competitive binding assay was developed using cloned porcine adenopituitary GHRH receptors expressed in human kidney 293 cells. Specific binding of [His’,‘% Tyr”‘,Nle”]hGHRH( 1-32)-NH? increased linearly with protein concentration (lo-45 pg protein/tube). Binding reached equilibrium after 90 min at 30°C and remained constant for at least 240 min. Binding was reversible to one class of high-affinity sites (& = 1.04 t 0.19 nM, B,,, = 3.9 2 0.53 pmol/mg protein). Binding was selective with a rank order of affinity (IC,) for porcine GHRH (2.8 -t 0.51 WV), rat GHRH (3.1 2 0.69 nM), [N-Ac-Tyr’,o-Arg’]hGHRH(3-29)-NH, (3.9 rf: 0.58 n144), and [DThr7]GHRH(1-29)-NH2 (189.7 + 14.3 nM), consistent with their binding to a GHRH receptor. Nonhydrolyzable guanine nucleotides inhibited binlding. These data describe a selective and reliable method for a competitive GHRH binding assay that for the first time utilizes rapid filtration to terminate the binding assay.

growth hormone-releasing hormone (GHRH) binding to cloned porcine GHRH receptor. PEPTIDES

Cloned receptor GHRH receptors

GHRH binding

Porcine GHRH receptor

to Dr. Mark L. Heiman,

hormone

(GHRH)

competitive binding assay for studying GHRH SAR using a rapid filtration method to terminate the binding assay.

HUMAN growth hormone-releasing hormone (hGHRH) has been suggested as a therapy for growth disorders [for review see (11,22)]. Structure-activi,ty relationships (SAR) of hGHRH have been studied for more than a decade. Information has been obtained from studying analogue stimulation of GH secretion in vivo and in vitro (7). Although extensive GHRH analogue libraries exist, several technical problems are encountered when studying the SAR for GHRH binding to its receptor. Like other peptides, stability of GHRH ligands in the presence of tissue homogenates represents the foremost difficulty to establish competitive binding assay. Degradation of GHRH by plasma (10) and tissue (3) proteases, as well as by nonenzymatic means (2,9), is reported to reduce GHRH bioactivity. Another technical problem inherent to most peptides and the GHRH molecule is the strong adherence of GHRH to glass and plastic. Finally, it is difficult to locate a tissue rich in GHRH receptors. Recently, we cloned the porcine GHRH receptor (15). This receptor was found to have 86%, 82%, and 80% homology to man, rat, and mouse GHRH receptor, respectively. The objective of this study was to characterize GHRH binding to this receptor. We report the development of a sensitive, selective, and reliable

’ Requests for reprints should be addressed 0540, Indianapolis, IN 4628.5.

Growth hormone-releasing

METHOD

Materials

Polyethylenimine (PEI), bacitracin, trypsin, and phenylmethyl-sulfonyl fluoride (PMSF) were from Sigma (St. L.&s, MO). Bovine serum albumin (BSA) and fetal bovine serum were from Intergen (Purchase, NY). Leupeptin was purchased from Protein Research Foundation (San Francisco, CA). Lima bean trypsin inhibitor (LBI) was from Worthington (Freehold, NJ). Minimum essential medium (s-MEM) and cell culture freezing mediumDMSO were from GIBCO-BRL (Grand Island, NY). Glucagon, vasoactive intestinal peptide (VIP), and secretin were from Bachem (Torrance, CA). Guanine nucleotide mono-, di-, and triphosphate (GMP, GDP, GTP, respectively), guanylyl-imidodiphosphate (GMPNP), guanylyl(&y-methylene)-diphosphonate (GMPPCP), and guanosin-S-0-(34iotriphosphate) (GTP-7-S) were from Boehringer Mannheim (Indianapolis, IN). [DThr7]GHRH( l -29)~NH, was a kind gift from Dr. David H. Coy

Division

1469

of Endocrinology,

Eli Lilly and Company,

Corporate

Center, Drop

HASSAN ET AL.

1470

TABLE 1

Binding Assay

DEGRADATION OF GHRH BY CELL HOMOGENATE Area Under the Curve (UV Absorbance at 214 nm) Time of Incubation (mitt)

Without Inhibitors

With Inhibitors

0 60 120

26.10 t 0.38 23.61 ? 0.36 18.22 + 2.02

25.87 2 0.87 25.16 2 0.93 25.51 c 0.35

Values are mean 4 SEM of three replicates. Membrane homogenate was incubated at 30°C with [His’, Nle27]hGHRH(l-32)-NH? in the presence or absence of enzyme inhibitors for the time indicated. Intact peptide was analyzed using reverse-phase HPLC.

(Tulane University School of Medicine, New Orleans, LA). All other GHRH analogues were synthesized using solid-phase methodology. L&and and Radiolabeling [His’,Nlez7]hGHRH( 1-32)-NH2 was radioiodinated using the chloramine-T method as described by Seifert et al. (19). Separation of “‘1-Na from radiolabled ligand was performed with a disposable Cl8 column (Sep-Pak, Waters). Further purification to obtain the monoiodinated species was performed by reversedhase HPLC (Cl8 column, Vydac, 0.46 X 25 cm, pore size 300 x, 27% to 32% CHXCN, 0.1% trifluoroacetate (TFA) in 75 min at 1 ml/min). Fractions corresponding to monoiodinated hormone were isolated and used only for 2 weeks. Average specific activity was 2000 Ci/mmol(550 @/mg). Ligand was diluted in 50% CH$ZN in 0.1% TFA to a final concentration of 0.08 nM. Binding affinities for the radiolabeled ligand, its nonradioactive monoiodinated homolog, and its parent [His’,Nle”]hGHRH( 1-32)NH2 for the rat anterior pituitary GHRH receptor are similar (19).

Reaction vials used in the binding assay were 12 x 75 mm Baxter borosilicate glass tubes. Binding was carried out in a final volume of 300 ~1 comprised of binding buffer, membrane homogenate (50 ~1; 30 hg protein), and labeled ligand (5-10 ~1; 0.08 nM) at 30°C for 120 min. Binding in the presence of 1 @Z unlabeled ligand (10 ~1) was defined as nonspecific binding. Because GHRH tends to stick to plastic and glass, all dilutions of labeled and unlabeled peptides were achieved using 50% CH3CN in volumes ranging from 10 to 20 ~1 (1.5-3.5% CH$N). Binding buffer used was the same as the homogenization buffer but supplemented with 0.25% BSA. To stop the binding reaction, 5 ml ice-cold binding buffer was added to each tube. Bound ligand was separated from free ligand by rapid (< 15 s) filtration through glass-fiber filters (Whatman GF/B) that had been pretreated for at least 2 h with 1.2% PEI in HEPES buffer. Subsequently, filters were washed three times with 10% CHJN in HEPES buffer. GHRH Stability Under standard conditions, 10 @f of [His’,Nle*‘]hGHRH(l32)-NH2 was incubated with membrane homogenate (210 ,ug protein) in the presence or absence of enzyme inhibitors for 0, 60, and 120 min. Reaction was stopped by removing 500 ~1 of incubation buffer and adding 12.5 1.11of 1 M TFA. The mixture was then perifused through an activated disposable Cl8 column (Sep-Pak, Waters), washed with 0.1% TFA, and the bound GHRH was eluted by 50% CH,CN. Fractions were collected, dried, reconstituted in 0.1% TFA, and analyzed using reversehase HPLC (Cl8 column, Vydac, 0.46 x 25 cm, pore size 300 x 24-42% CHJJN, 0.1% TFA in 25 min at 1 ml/min). Peptide re’coveries were determined by the integrated area under the peak representing the parent peptide (Table 1).

Cell Culture

Adenovirus-transformed human kidney 293 cells, stably transfected with cDNA plasmid (pRC/CMV) containing porcine GHRH receptor cDNA, were maintained as described previously (15). Cells were grown to confluency, then harvested using a rubber policeman after incubation (5 min at 37°C) with 1% trypsin and 0.2% EDTA in s-MEM solution. Cells were washed with ice-cold s-MEM and recovered by centrifugation (400 x g for 10 min at 4°C). Finally, cells were pooled to a concentration of 100 million cells/ml in cell culture freezing media and stored at -70°C until processed for binding. Cell Homogenate Tissue Concentration, pg protein/tube

Frozen cells were thawed at 30°C washed three times with sMEM, and recovered by centrifugation at 400 X g for 10 min at 4°C. Cells were then suspended in 3 ml ice-cold homogenization buffer (50 mM HEPES, 7 m&4 MgC&, 5 mM EDTA, 50 &ml PMSF, 50 @ml bacitracin, 50 fig/ml lima bean trypsin inhibitor, and 20 &ml leupeptin; pH 7.4) and incubated on ice for 10 min. Swollen cells were then homogenized using Dounce Teflon-glass homogenizer (10 strokes, 500 rpm). The homogenate was centrifuged (400 X g at 4°C) for 10 min. The pellet was discarded; the supematant protein concentration was measured as described earlier (4) and was further diluted to 600 pg protein/ml for binding assay.

FIG. 1. Specific [His’,‘?’ I-Tyr’“,Nle*7]hGHRH( 1 -32)-NH2 binding to cloned porcine GHRH receptor as a function of protein concentration. Binding assay was carried out at final volume of 300 ~1 binding buffer, membrane homogenate (50 ~1; lo-60 pg protein), and labeled ligand (5- 10 ~1; 0.08 nA4) at 30°C for 120 min. Bound ligand was separated from free ligand by filtration through glass-fiber filters (GF/B), which have been pretreated for 2 h with 1.2% polyetbylenimine in HEPES buffer. Specific binding was calculated as the difference between total binding (tracer alone) and nonspecific binding (tracer in presence of 1 @4 [His’,Nle*‘]hGHFW( 1-32)-N&}. Each point represents the average of triplicate determinations. These data are representative of five independent experiments.

1471

GHRH RECEPTOR BINDING CHARACTERISTICS

18r

k1=0.02 nM-‘.min-’

T

01........““““““““” 0

60

120

180

240 Time,

300

360

420

1

480

min

FIG. 2. Binding kinetics of [His’,Nle*‘]hGHBH(l-32)-NH, binding to cloned porcine GHRH receptor. Binding was performed as stated in Fig. 1 legend with 30 ,ug protein/tube for the time periods indicated. At 120 min. dissociation reaction was initiated by adding 1 @f of [His’,Nlen]hGHFUI( 1-32)-N&. Experimental data were analyzed using computer program INPLOT. Each point represents the mean k SEM of triplicate determinations. These data are representative of five independent experiments.

5

Analog Concentration,

M

FIG. 3. Inhibition of [His’,‘= I-Tyr”‘,NleZ7]hGHBH( l -32)-NH, binding to cloned porcine GHRH receptor by different GHRH analogues. Binding was performed as stated in Fig. 1 legend with 30 pg protein/tube. Experimental data were analyzed using computer program LIGAND. Each point represents the average of triplicate determinations. These data are representative of five independent experiments.

Data Analysis

DISCUSSION

LIGAND (18) was used to calculate equilibrium dissociation constant (&) and maximum concentration of binding sites (B,,,). INPLOT (GraphPad Software; San Diego, CA) was used to calculate association and dkociation rate constants (K, and K-,), and ALLFIT (8) was used to calculate concentrations of peptides that inhibit 50% of maximum GHRH binding (I&). RESULTS

[His’,Nle*‘]hGHRH( l -32)-NH, is stable in the presence of cell homogenate and protease inhibitors for at least 120 min (Table 1). Specifically bound radioligand was linearly related to total protein concentration (Fig. 1) and 30 pg protein/tube was used in each of the following experiments. Specific binding increased with time, reached equilibrium after 90 min, and was maintained at equilibrium for at least ;!40 min (Fig. 2). Specific binding was 87% of total binding. Binding was reversible (Fig. 2) and dissociated by first-order kinetics with T,n of 5.28 -C0.33 h and K_., = 0.0411 + 0.004 mm’. Isotherm binding data (Fig. 3, closed circles) indicated one class of high-affinity binding sites (Kd = 1.038 + 0.19 IlM, It,,, = 0.39 2 0.06 r&l). This corresponds to 3.9 pmol/mg protein (35,000 receptors/cell). Competition studies with [His’,Nle*‘]hGHRH( I.-32)-NH, (Fig. 3) indicated that KS0 was 2.06 2 0.21 n&f. Computed I&, for both rat and porcine GHRH was 3.1 + 0.69 n&l and 2.8 + 0.51 nM, respectively. [DThr’]GHRH( l -29)~NH,, an analogue reported to have low biological activity (13), was a significantly, F(5, 17) = 4.34, p < 0.01, less potent competitor (IC, = 189.7 + 14.3 nkf) of [His’,‘= I-Tyr’“,Nle27]hGHRH( l -32)-NH, binding (Fig. 3). A potent GHRH antagonist [N-Ac-Tyr’,o-Arg]hGHRH(3-29)NH2 (6) bound to the receptor (Fig. 3) with high affinity (ICso = 3.9 + 0.58 n&f). Several peptides with significant homology to GHRH (VIP, glucagon, and secretin) did not compete with bound labeled ligand (Fig. 4). Binding of GHRH to its receptor is reported to be dependent on guanine nucleotide regulatory proteins. Indeed, we observed such inhibition. Both GDP and GTP inhibited GHRH binding by 62% and 59%, respectively (Fig. 5). Nonhydrolyzable analogues GTP-y-S and GMPPNP resulted in approximately 50% and 56% decrease in GHRH binding, respectively. Addition of GMP did not alter binding.

Physicochemical characteristics of GHRH render binding studies of the neuropeptide problematic. Several technical problems have been described previousiy (1,20,21). In the present study, we solved these technical obstacles and characterized binding of [His’,‘” I-Tyr’“,NleZ7]hGHRH( l -32)-NH, to cloned porcine GHRH receptor. Nonspecific binding presents a serious problem. Struthers, Perk, and Vale (21) avoided such problem by performing dilutions and transfers in 50% acetonitrile, which was eliminated by lyophilization prior to assay. We found dilutions in 50% acetonitrile absolutely necessary. However, we also observed that concentrations (up to 5%) of this solvent do not interfere with equilibrium binding up to 240 min (data not shown). Thus, we diluted and transferred peptides in 50% acetonitrile. We observed considerable binding and apparent displaceable binding to almost all test tubes examined. Under the binding assay conditions and in the absence of tissue homogenate, 66%,

20-

&i

,

,,,L9

"."'I 1o-g 10-'0 1"""

10-e

Peptidc

. .~,.~ 10-s

10-7

Concentration,

10-S

M

FIG. 4. Inhibition of [His’,‘*“I-Tyr’O, Nle27]hGHRH(l -32)-NH2 binding to porcine GHRH receptor by related peptides. Binding was performed as stated in Fig. 1 legend with 30 pg protein&&e. Each point represents the average of triplicate determinations. These data are representative of five independent experiments.

1472

HASSAN ET AL.

20

Nucleotide,100 PM

FJG. 5. Nonhydrolyzable guanyl nucleotides reduced [His’,“‘ITyr”‘,Nle*‘]hGHRH(l-32)-NH, binding to cloned porcine GHRH receptor. Binding was performed at conditions stated in Fig. 1 legend with 30 pg protein/tube, and addition of GMP (guanosin-5’-monophosphate), GMPNP (guanylyl-imidodiphosphate), GMPPCP [guanylyl@,y-methylene)-diphosphonate], GDP (guanosine-5’-diphosphate), GTP (guanosine-5’-triphosphate), or GTP-y-S [guanosin-5’-O-(3-thiotriphosphate)]. Each bar represents the mean 2 SEM of triplicate determinations. These data are representative of five independent experiments.

65%, 64%, 53%, and 27% of total counts added adhere to Teflon, polypropylene, polystyrene, flint glass, and borosilicate glass tubes, respectively. Moreover, 50% of these counts were displaced by addition of nonlabeled ligand to Teflon, polypropylene, and polystyrene tubes. The least competitive binding in the absence of membranes was observed using borosilicate glass tubes, whereas the greatest binding was found to plastic tubes. Indeed, Seifert et al. (20) used borosilicate glass test tubes as a reaction vial, but they and other investigators (1,13,17) used plastic tubes and centrifugation to separate bound from free ligand. We were concerned about using plastic centrifuge tubes; thus, we used filtration through borosilicate glass fiber filters that were presoaked with 1.2% PEI for 2 h. We rapidly washed trapped bound ligand with 10% acetonitrile in HEPES to eliminate free radioligand. This eliminated nonspecific binding to filters. Abribat, Boulanger, and Gaudreau (1) reported significant high-affinity

binding to such filters in the absence of pituitary homogenate. They presoaked the filters with 0.1% PEI; however, they evidently did not wash with buffer containing acetonitrile. The labeled ligand used in our studies was prepared and purified according to Seifert et al. (19) with one exception. Rather than dilute purified tracer with aqueous buffer and store at -20°C we stored the concentrated labeled ligand at 4°C in solvent resulting from HPLC purification. For assay, [His’,12?Tyr”,Nle”]hGHRH( l -32)-NH, was diluted in 50% acetonitrile and added directly (5 - 10 ~1) to reaction tubes. Others ( 1,17) have purchased [‘251-Tyr’o]hGHRH(1-44)-NH, to probe GHRH receptors. Our ligand was stable for at least 2 h in HEPES buffer containing EDTA and a cocktail of enzyme inhibitors that inhibit trypsin, chymotrypsin, metalloproteases, plasmin, papain, cathepsin D, exo-peptidases, and endo-peptidases (Table 1 and Fig. 2). Others (1) have found that EDTA was sufficient to inhibit GHRH degradation by pituitary homogenates. Binding of labeled ligand to membrane preparations containing pGHRH receptor transfectants was of high affinity (& = 1.038 nM). This is similar to that reported for rat pituitary cells (21), and COS cells transfected with the human GHRH receptor (13), but lower than that calculated for kidney 293 cells transfected with human GHRH receptor cDNA (17). Such discrepancy may reflect differences in radioligand, G-protein coupling, and equilibrium binding parameters. In accord with published literature (16), the analogue [D-Thr7]hGHRH( 1-29)~NH2 was found to have low binding affinity. Occupation of G-protein-coupled receptors by agonists results in an increase in the rate of exchange of GTP for GDP, with subsequent dissociation of the cr and/Q subunits of the G-protein heterotrimer. This coincides with a decrease in the affinity of the receptor for agonist ligands (5). Such a decrease in affinity of GHRH receptor for its agonist can be observed by addition of nonhydrolyzable GTP analogues ( 1,12,2 1). Nonhydrolyzable analogues GTP-y-S and GMPPNP resulted in approximately 50% and 56% decrease in GHRH binding, respectively. Addition of GMP did not alter binding. These data are similar in magnitude to that observed by others (1,12,20). In conclusion, we characterized a reliable radioreceptor assay for GHRH. Our methodology will facilitate study of the threedimensional stoichiometry of GHRH-receptor interaction and can be used to discover new molecules that bind to the GHRH receptor. ACKNOWLEDGEMENTS

We are grateful to Dr. D. H. Coy for providing the [D-Thr’]GHRH( l29)-NH* analogue. We also appreciate the technical assistance of G. C. Harris, P. L. Surface, and D. Ross. The authors wish to thank Drs. Robert F. Bruns and David L. Nelson for the critical review of this manuscript. Results from this research were presented in part at the 76th Annual Meeting for the Endocrine Society, Anaheim, CA, Abstract #669.

REFERENCES Abribat, T.; Boulanger, L.; Gaudreau, P. Characterization of [‘*‘ITyr’o]-human growth hormone-releasing factor (l-44) amide binding to rat pituitary: evidence for high and low affinity classes of sites. Brain Res. 528:291-299; 1990. Bongers, J.; Heimer, E. P.; Lambros, T.; et al. Degradation of aspartic acid and asparagine residues in human growth hormone-releasing factor. Int. J. Pept. Protein Res. 39:364-374; 1992. Boulanger, L.; Lazure, C.; Lefrangois, L.; Gaudreau, P. Proteolytic degradation of rat growth hormone-releasing factor (l-29)

amide in rat pituitary and hypothalamus. Brain Res. 616:39-47; 1993. 4. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. B&hem. 72:248-254; 1976. of receptor/GTP-binding protein inter5. Cerion, R. A. Reconstitution actions. Biochim. Biophys. Acta 1071:473-501; 1991. 6. Coy, D.; Hocart, S. J.; Murphy, W. A. Human growth hormonereleasing hormone analogues with much improved in vitro growth

GHRH RECEPTOR BINDING CHARACTERISTICS

7.

8.

9.

10.

11. 12.

13.

14.

hormone-releasing potencies in rat pituitary cells. Eur. J. Pharmacol. 204:179-185; 1991. Coy, D. H.; Murphy, W. A ; Lance, V. A.; Heiman M. L. Differential effect of N-terminal modifications on the biological potencies of growth hormone releasing factor analogues with varying lengths. J. Med. Chem. 30:219-222; 1986. De Lean, A.; Munson, P. J.; Rodbard, D. Simultaneous analysis of families of sigmoidal curves: Application to bioassay, radiohgand assay, and physiological dose-response curves. Am. J. Physiol. 235(2):E97-E102; 1978. Friedman, A. R.; Ichhpurani, A. K.; Brown, D. M.; et al. Degradation of growth hormone releasing factor analogs in neutral aqueous solution is related to deamidation of asparagine residues. Int. J. Pept. Protein Res. 37: 14-20; 1991. Frobman, L. A.; Downs, T. R.; Heimer, E. P.; et al. Dipeptidylpeptidase N and trypsin-lie enzymatic degradation of human growth hormonereleasing hormone in plasma. J. Clin. Invest. 83:1533-1540, 1989. Frohman, L. A.; Jansson, J. 0. Growth hormone releasing hormone. Endocr. Rev. 71223-253; 1986. Gayhnn, B. D.; Charles, E. L.; Zysk, J. R.; Clarke; I. J. Photoaffinity cross-linking to the pituitary receptor for growth hormone releasing factor. Endocrinology 135:950-955; 1995. Gaylinn, B. D.; Harrison, .J. K.; Zysk, J. R.; et al. Molecular cloning and expression of a hum.an anterior pituitary receptor for growth hormone-releasing hormone. Mol. Endocrinol. 7:77-84; 1993. Heiman, M. L.; Nekola, M. V.; Murphy, W. A.; et al. An extremely sensitive in vitro model for elucidating structure-activity relation-

15.

16.

17.

18.

19.

20.

21.

22.

ships of growth hormone-releasing factor analogs. Endocrinology 116:410-415; 1985. Hsiung, H. M.; Smith, D. P.; Zhang, X. Y.; Bennett, T.; Rosteck, P. R.; Lai, M. H. Structure and functional expression of a complementary DNA for porcine growth hormone-releasing hormone receptor. Neuropeptides 25: 1 - 10; 1993. Lefratqois, L.; Gaudreau, P. Identification of receptor-binding pharmacophores of growth-hormone-releasing factor in rat adenopituitary. Neuroendocrinology 59:363-370; 1994 Mayo, K. E. Molecular cloning and expression of a pituitary-specific receptor for growth hormone-releasing hormone. Mol. Endocrinol. 6:1734-1744; 1992. Munson, P. I.; Rodbard, D. Ligand: A versatile computerized approach for characterization of ligand-binding systems. Anal. Biochem. 107:220-239; 1980. Seifert, H.; Perrin, M.; Rivier, J.; Vale, W. Binding sites for growth hormone releasing factor on rat anterior pituitary cells. Nature 313:487-489; 1985. Seifert, H.; Perrin, M.; Rivier, J.; Vale, W. Growth hormone-releasing factor binding sites in rat anterior pituitary membrane homogenates: Modulation by glucocorticoids. Endocrinology 117:424-426; 1985. Struthers, R.; Perrin, M.; Vale, W. Nucleotide regulation of growth hormone-releasing factor binding to rat pituitary receptors. Endocrinology 124:24-29; 1989. Thomer, M. 0. On the discovery of growth hormone-releasing hormone. Acta Paediatr. Suppl. 388:2-7; 1993.