0013.7227/92/13311-0’L68$03.00/0 Endocrinology Copyright vZ 1992 by The Endocrine

Vol. 131, No. 1 Printed in U.S.A.

Society

Identification of Hormone-Binding Regions of the Luteinizing Hormone/Human Chorionic Gonadotropin Receptor Using Synthetic Peptides* PATRICK

C. ROCHE,

ROBERT

J. RYAN,

Departments of Laboratory Medicine/Pathology D. J. McC.), Mayo Clinic, Rochester, Minnesota

AND

DANIEL

J. MCCORMICK

(P.C.R.) and Biochemistry/Molecular 55905

Biology (R.J.R.,

ABSTRACT A comprehensive series of overlapping synthetic peptides have been used to study the relationship between the primary structure of the ovarian receptor for LH/human CG (hCG) and hormone binding. Twenty-four consecutive, overlap peptides that replicate the entire extracellular domain of the rat luteal receptor have been synthesized by standard solid-phase techniques on an automated synthesizer. Eight additional peptides from the extracellular domain and three peptides replicating the putative extracellular loop regions have also been synthesized. Each peptide was evaluated in RRAs for interaction with hCG by measuring its ability to competitively inhibit binding of l”IhCG to membrane receptor. Twelve peptides were found to be potent in RRAs and caused a reduction of half-maximal binding of ““I-hCG at concentrations of lo-250 x lo-” M. The 12 active peptides (and adjacent inactive peptides) defined at least 4 independent receptor regions that can interact with hormone. One site near the NH2-

terminus was localized to receptor residues Arg”-Pro”‘. Two more sites of hormone interaction were identified by peptides replicating residues Arg’w’.Thr1’5 and Tyf’5”-phe’7’. A fourth binding region was identified in the third putative extracellular loop, replicated by rat luteal receptor peptide Ly?“-Lys?. The amino acid sequences of the four active rat LH/hCG receptor regions were aligned and compared with published sequences for other glycoprotein hormone receptors. Three regions (Argl(‘-Thrl”, Tyr’““-Phez7’, and LysS7”-Ly?) showed high sequence homology with the human LH/hCG receptor, human TSH receptor, and rat FSH receptor and may represent contact sites for the a-subunit of hormone. The other binding region, Arg”-Pro3’ had low sequence homology with the other glycoprotein hormone receptors and is postulated to be a binding determinant for p-hCG/LH. This report demonstrates that synthetic overlap peptides of confirmed sequence can be used to successively identify hormone interaction sites of glycoprotein hormone receptors. (Endocrinology 131: 268-274, 1992)

T

tural organization of the rat LH receptor gene and have determined that the coding region contains 11 exons and 10 introns. Exons l-10 code for the NH2-terminal half of the molecule whereas exon 11 codes for the COOH-terminal half. Exon 11 and the preceding 10th intron are proposed to be the homologue of the intronless gene of those G-protein coupled receptors that bind small ligands (e.g. &adrenergic receptors). The LH/hCG, FSH, and TSH receptors have a large NH,terminal half (341-418 residues) that is modeled to be extracellular and to contain binding sites for the sizable (28-38 kilodaltons) glycoprotein hormones (10-13). In contrast, those G protein-coupled receptors that bind small ligands are believed to do so through transmembrane regions. The glycoprotein hormone receptors thus constitute a unique subclass of G protein-coupled receptors. The extracellular domain of the LH/hCG receptor contains six putative sites for N-linked glycosylation (Asn-X-Ser/Thr) in the rat, pig, and human (3, 4, 14). Multiple leucine-rich segmentswith sequence similarity to a group of leucine-rich glycoproteins are also found in the extracellular domain (3). The leucinerich glycoproteins participate in protein-protein interactions (15, 16), possibly by the formation of amphipathic helices (16). In the LH/hCG receptor, the 10 exons that code for the NHz-terminal half display both DNA and amino acid sequence similarities with each other, and an exon-dependent repetitive motif has been proposed by Koo et al. (8). In addition to full-length cDNAs for the LH/hCG receptor,

HE BIOCHEMICAL and physiological actions of LH and human CG (hCG) are initiated by binding of circulating hormone to specific, high affinity receptors in the plasma membrane of target cells in gonadal tissue (1). LH/hCG receptors are coupled to intracellular effector systems, most notably adenylate cyclase, through guanine nucleotide-binding regulatory proteins or G-proteins (2). Recent clonings of complete complementary DNAs (cDNAs) for LH/hCG receptors (3, 4), and for receptors to the other glycoprotein hormones, FSH [follitropin (5)] and TSH [thyrotropin (6, 7)], have revealed a high degree of homology (approximately 60%) in deduced amino acid sequence. The COOH-terminal half of these proteins are also homologous to the larger family of G protein-coupled receptors represented by adrenergic receptors and rhodopsin. A characteristic feature of this receptor family is the presence of seven putative membrane-spanning domains with the NHz-terminus of the molecule positioned extracellularly and the COOH-terminus oriented intracellularly. Receptors for the glycoprotein hormones also contain seven segments of hydrophobic amino acids in their COOH-terminal regions and are thought to be similarly oriented in the plasma membrane. Koo et al. (8) and Tsai-Morris et al. (9) have recently published the strucReceived January 31, 1992. Address all correspondence and requests for reprints Roche, Department of Laboratory Medicine/Pathology, Rochester, Minnesota 55905. * This work was supported by NIH Grants HD-22735 the Mayo Foundation, and the Mellon Foundation.

to: Patrick C. Mayo Clinic, and HD-9140,

268

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.

HORMONE-BINDING

REGIONS

truncated variants that presumably result from alternative messengerRNA splicing have been identified in porcine and rat testis cDNA libraries (4, 11). These receptor forms contain the entire extracellular domain but lack the putative transmembrane regions. A secreted, soluble receptor form (potential antagonist) is therefore possible. Expression in mammalian cells of truncated rat ovarian receptor, containing signal peptide and extracellular domain, resulted in a high affinity hCG binding protein that was trapped within the cells (10, 12). However, Tsai-Morris et al. (11) reported that expression of a truncated testis cDNA in COSl cells did result in production and secretion of a soluble hCG binding protein. Knowledge of the primary amino acid sequence has provided us with the opportunity to implement exciting strategies for study of the LH/hCG receptor system utilizing synthetic peptides. The utility of synthetic peptides in locating functional domains of the glycoprotein hormones is well documented by our laboratory (17-19, 20-22). A synthetic peptide approach has also been used by other laboratories to study not only glycoprotein hormone-receptor interactions (23-25), but also to define ligand-receptor interactions involving coagulation factors (26), lymphocyte (27), and platelet adhesion (28), and cholinergic binding sites (29, 30). In this report we have used a comprehensive series of overlap and nested peptides encompassing the entire extracellular domain (i.e. 341 residues)of the rat luteal LH/hCG receptor to identify specific sequences that participate in hormone binding. Materials

and Methods

Peptide synthesis A consecutive series of 24 overlapping peptides to the 341 amino acid extracellular domain of the rat ovarian LH/hCG receptor were synthesized as COOH-terminal amides by standard solid-phase techniques on 0.5 mmol p-methylbenzhydrylamine resin using an automated peptide synthesizer (Applied Biosystems 430A, Foster City, CA). The series of overlap peptides, as shown in Fig. 1, were 20 residues in length. The NH*-terminal and COOH-terminal ends of each peptide overlapped with adjacent neighboring peptides by 6 amino acids (except for peptide Ala322-Arg341 which overlapped its preceding peptide by 7 residues). Three peptides replicating the putative extracellular loop regions, were also synthesized (Fig. 1). Note that peptide LYP-LYS~*~ has a tyrosine (Y) added to its COOH-terminus for possible radioiodination. To further define active regions identified with the consecutive overlap series, 7 additional nested peptides from the extracellular domain (Fig. 1) were synthesized. Peptide A~p’~-Lys*’ was made due to the low solubility of the original overlap peptide, Ala7i-Leu9’. Completed peptides (0.5 mmol) were removed from the resin using liquid hydrogen fluoride (0 C, 1 h) with 10% anisole and 3% dimethyl sulfide. The peptides were then purified by reverse-phase HPLC (Beckman Instruments, Palo Alto, CA) on a preparative C-18 column (Vydac) in a trifluoroacetic acid/acetonitrile solvent system. Peptide containing fractions were pooled, lyophilized, and stored with a dessicant in the dark at room temperature. Peptide homogeneity was confirmed by microsequencing (10 nmol) on a gas-phase automated protein sequentator (Applied Biosystems 470A), and the resultant phenylhydantoin derivatives identified by reverse-phase HPLC using an on-line PTH analyzer (Applied Biosystems 120A). Peptide solutions were kept mildly acidic (pH 4-5) in order to prevent or minimize polymerization by disulfide bond formation,

OF THE LH/hCG Radioreceptor

RECEPTOR

269

assays

Peptides were evaluated for their ability to competitively inhibit binding of I*?-hCG to particulate (membrane bound) rat ovarian LH/ hCG receptor as previously described (17). Membrane fractions from superovulated rat ovaries and ?hCG were prepared as previously described (31). Solutions of purified peptides (10 mg/ml) in distilled water were made fresh for each assay from lyophilized material. Two nanograms of radiolabeled hCG were incubated overnight (16 h) at 20 C with 1.25 mg membrane (wet weight) and graded doses of peptide or unlabeled hormone. Initial experiments compared the effects of a 2-h preincubation of radiolabeled hCG with receptor peptides before addition of ovarian membrane and found that the order of addition did not increase or decrease the potency of receptor peptides. Nonspecific binding was measured by the addition of a large excess (loo-fold) of unlabeled hCG, and the data shown reflect binding obtained by subtracting the nonspecific component from total binding. The inhibitory potency (I&) of each peptide was expressed as the molar concentration causing half-maximal inhibition of iz51-hCG binding and was calculated from linear regression analysis of dose response curves. Amino acid sequence alignment was performed with software from Genetics Computer Group (32).

Results

The effectiveness of synthetic LH/hCG receptor (LHR) peptides to competitively inhibit lz51-hCG binding to rat luteal membranes provides an indirect measure of peptide interaction with hCG. A total of 35 receptor peptides were evaluated in RRAs (Fig. 1) and 12 LHR peptides had inhibitory activity. The potency of synthetic peptides, expressedas the molar concentration causing half-maximal inhibition of binding of ‘251-hCG, is summarized in Table 1. The doseresponse curves for 10 active LHR peptides (9 from the extracellular domain and 1 from the proposed third extracellular loop) are displayed in the graphs of Fig. 2. Six LHR peptides near the NH*-terminus had strong inhibitory activity with potencies of Arg*‘-Pro3* > Arg21-Va141 > Ala’5-Pro38 > Ala’5-Leu34> Leu29-Arg48> Arg31-Arg48.Adjacent overlap peptides Pro’-Arg*‘, and Pro43-Asp62had no inhibitory activity. The data generated with this nested group of peptides localizes one hormone-binding region in the LH/hCG receptor to residuesArg2’-Pro38. LHR peptides Asn99-Asp”8 and Arg102-Thr”5 also inhibited hCG binding to luteal receptor, and identified a second independent receptor site that can associatewith hormone. Since adjacent overlap peptides for Asn99-Asp”8 were without activity, the hormone-binding site is maximally debmited within residues Arg”*-Thr’15. Overlapping peptides Ser239-Cys258 and Tyr253-Phe272 defined a third binding region. The nested peptide Tyr253-Lys266 displayed one-third the potency of Tyr253-Phe272, and therefore this region is maximally delimited by the latter peptide. The final peptide with inhibitory activity was Lys573-Lys583 (Y). This peptide replicated the third putative extracellular loop and identified a fourth region of the receptor with the potential to associatewith hormone. Sequence alignments of the four active regions of the rat LH/hCG receptor with homologous sequencesof the human LH/hCG receptor, the human TSH receptor and the rat FSH receptor are shown in Fig. 3.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.

HORMONE-BINDING

270

SEQUENCE OF rLH

RECEPTOR

Overlap l-20

No 1

PEPTIDES

Arg-Glu-Leu-Ser~Gly-Ser-Arg-Cys-Pro-Glu-Pro-Cys-Asp-Cys-Ala-Pro-Asp-Gly-Ala-Leu Ala-Pro-Asp-Gly-Ala-Leu-Arg-Cys-Pro-Gly-Pro-Arg-Ala-Gly-Leu-Ala-Arg~Leu-Ser-Leu

29-48 43-62

Leu-Ala-Arg-Leu~Ser~Leu-Thr-~r-Leu-Pro-Val-Lys-Val-Ile-Pro-Ser-Gln-Ala-Phe-Arg

57-16

Glu-Ile-Ser~Gln~Ser-Asp-Ser-Leu-Glu-Arg-Ile-Glu-Ala-Asn-Ala-Phe-Asp-Asn-Leu-Leu

71-90

Ala-Phe~Asp~Asn-Leu-Leu-Asn-Leu-Ser-Glu-Leu-Leu-Ile-Gln-Asn-Thr-Lys-Asn-Leu-Leu

99-118

Voll31.

Peptides

15-34

85-104

Endo.

REGIONS OF THE LH/hCG RECEPTOR

Pro-Ser-Gln-Ala-Phe-Arq-Gly-Leu-Asn-Glu-Asn-Glu-Val-Val-Ly6-Ile-Glu-Ile-Ser-Gln-Ser-Asp

Asn-Thr-Lys-Asn-Leu~Leu-Tyr-Ile-Glu-Pro-Glu-Pro-Gly-Ala-Phe-Thr-Asn-Leu-Pro-Arg-Leu-Lys Asn-Leu-Pro~Arg~Leu-Lys-Tyr-Leu-Ser-Ile-Cys-Asn-Thr-Gly-Ile-Arg-Thr-Leu-Pro-Asp

113-132

Ile-Arg-Thr-Leu-Pro-Asp-Val-Thr-Lys-Ile~Ser-Ser-Ser-Glu-Phe-Asn-Phe-Ile-Leu-Glu

127-146

Phe-Asn-Phe-Ile-Leu-Glu-Ile-Cys-Asp-Asn-Leu~His-Ile-Thr-Thr-Ile-Pro-Gly-Asn-Ala

141-160 155-174

Val~Thr-Leu-Lys-Leu-~r-Gly-AsnGly-Phe-G1u-Glu-Val-Gln-Ser-His-Ala-Phe-Asn-Gly

169-188

Ser-His~Ala-Phe-Asn-Gly-Thr-Thr-Leu-Ile-Ser-Leu-Glu-Leu-Lys-Glu-Asn-Ile-~r-Leu

Thr-Ile-Pro-Gly-Asn-Ala-Phe-Gln-Gly-Met-Asn-Asn-Glu-Ser-Val-Thr-Leu-Lys-Leu-~r

183-202

Lys~Glu-Asn-Ile-~r-Leu-Glu-LysMet-His-Ser-Gly-Ala-Phe-Gln-Gly-Ala-Thr-Gly-Pro

197-216

Cln~Gly~Ala-Thr-Gly-Pro-Ser-Ile-LeuAsp-Ile-Ser-Ser-Thr-Lys-Leu-Gln-Ala-Leu-Pro

211-230

Lys-Leu-Gln-Ala-Leu-ProSer-His-Gly-Leu-Gly-Leu-Glu-Ser-Ile-Gln-Thr-Leu-Ile-Ala-Leu-Ser

225-244

Thr-Leu-Ile~Ala~Leu-Ser-Ser-Tyr-SerLeuLys-Thr-Leu-Pro-Ser-Lys-Glu-Lys-Phe-Thr

239-258

Ser-Lys-Glu-Lys~Phe-Thr-Ser-Leu-Leu-Val~Ala~Thr-Leu-Thr-~r-Pro-Ser-His-Cys-Cys

253-272

Tyr-Pro-Ser-His~Cys-Cys-Ala-Phe-ArgAsnLeu-Pro-Lys-Lys-Glu-Gln-Asn-Phe-Ser-Phe

267-286

Glu-Gln~Asn-Phe-Ser-Phe-Ser-Ile-Phe-Glu-Asn-Phe-Ser-Lys-Gln-Cys-Glu-Ser-Thr-Val

281-300

Gln-Cys-Glu-Ser-Thr-Val-Arg-Lys-Ala~Asp-Asn-Glu-Thr-Leu-Tyr-Ser-Ser-Ala-Ile-Phe-Glu

295-314 309-328

Tyr-Ser-Ala~Ile-Phe-Glu-Glu-Asn-Glu-Leu-Ser-Gly-Trp-Asp-~r-Asp-~r-Gly-Phe-Cys Tyr-Asp-Tyr-Gly-Phe-Cys-Ser-Pro-LysThr-Leu-Gln-Cys-Ala-Pro-Glu-Pro-Asp-Ala-Phe

322-341

Ala-Pro-Glu-Pro~Asp-Ala-Phe-Asn-ProCys-G1r-Ala-Phe-Leu-Arg

Extracellular

Loop Peptides

397-417

Glu-Ser-Gln-Thr-Lys-Gly-Gln-Tyr-Tyr-~r-Asn-His-Ala-Ile-Glu-Trp-Gln-Thr-Gly-Pro-Gly-Cys

484-503

Ser-Asn-Tyr-Met-Lys-Val-Ser-Ile-Cys-Leu-Pro-Met-Asp-Val-Glu-Ser-Thr-Leu-Ser-Gln Lys-Val-Pro-Leu-Ile-Thr-ValThr-Asn-Ser-LysTyr

573-583(Y)

Nested 9-21 15-38

Peptides

Pro-Glu~Pro-Cys-Asp-Cys-Ala-Pro-Asp-Gly-Ala-Leu-Arg Ala-Pro~Asp-Gly-Ala-Leu-Arg-Cys-Pro-GlyPro-Arg-Ala-Gly-Leu-Ala-Arg-Leu-Ser-LeuThr-Tyr-Leu-Pro

21-38

Arg-Cys-Pro-Gly-Pro-Arg-Ala-Gly-Leu-Ala-Arg-Leu-Ser-Leu-Thr-~r-Leu-Pro

21-41

Arg-Cys-Pro-Gly-Pro-Arg-Ala-Gly-Leu-Ala-Arg-Leu-Ser-Leu-Thr-~r-Leu-Pro-Val-Lys-Val

31-48 102-115

Arg~Leu-Ser-Leu-Thr-Tr-Leu-Pro-Val-Lys-Val-Ile-Pro-Ser-Gln-Ala-Phe-Arg Arg-Leu-Lys-Tyr-Leu-Ser-Ile-Cys-Asn-Thr-Gly-Ile-Arg-Thr

253-266

Tyr-Pro-Ser-His~Cys-Cys-Ala-Phe-Arg-Asn-Leu-Pro-Lys-Lys

73-87

Asp-Asn-Leu-Leu-Asn-Leu-Ser-Glu-Leu-Leu-Ile-Gln-Asn-Thr-Lys

FIG. 1. Primary structure of synthetic peptides from extracellular loops. The numbers refer to the position between adjacent peptides. Note that peptide 573-583

the rat ovarian LH/hCG receptor that replicate the extracellular of residues according to McFarland et al. (3) and the underline has a tyrosine (Y) added to its COOH-terminus.

Discussion

The deduced amino acid sequencesfor the LH/hCG receptor and for receptors to the other glycoprotein hormones (TSH and FSH) have been used to create structural models that have a large (-340 amino acids) extracellular domain in

domain indicates

and the proposed regions of overlap

the NH*-terminal half, seven transmembrane regions that are separated by alternating intracellular and extracellular loops (three of each), and a relatively short intracytoplasmic COOH-terminal domain (3-8, 33). We have used a comprehensive seriesof overlap and nested peptides, replicating all the proposed extracellular portions of the rat LH/hCG recep-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.

HORMONE-BINDING TABLE binding

1. Potencies of LHR to rat luteal membranes

peptides

for inhibiting

Icso,” M Peptide

(mean

+ x

l-20 9-21 15-34 15-38 21-38 21-41 29-48 31-48 43-62 57-16 71-90 73-87 85-104 99-118 102-115 113-132 127-146 141-160 155-174 169-188 183-202 197-216 211-230 225-244 239-258 253-266 253-272 267-286 281-300 295-308 295-314 309-328 322-341 397-417 484-503 573-583(Y) ’ I& for unlabeled b NA = no inhibitory

REGIONS ‘““I-hCG

No. of assays

SE)

10-e

NAh NA 119 + 14.4 86.6 + 11.4 9.67 + 2.06 15.4 k 2.40 145 + 22.9 237 2 29.2 NA NA NA NA NA 66.0 + 12.9 40.1 rt 8.46 NA NA NA NA NA NA NA NA NA 138 f 15.6 129 * 12.1 39.8 + 4.52 NA NA NA NA NA NA NA NA 232 k 45.9

7 6 6 16 7 7 6 I

6 I

3

3

3 8 9 6 4 4 3 3 3 4

hCG = 4.58 f 0.43 x 10-l’ M. activity at highest dose tested

(500 X 1Om6 M).

tor, to define specific sequences that participate in the binding of hCG. Twelve active peptides identified a minimum of four discrete receptor regions as potential hormone-binding

OF THE

LH/hCG

271

domains. Three of the four regions were located in the NH,terminal half of the receptor, and a fourth was found in the proposed third extracellular loop. From the initial series of 24 overlap peptides, Ala”-Leu4 and Leu2”-Arg4H identified a binding site in the NHZ-terminal end of the receptor. The inactivity of overlap peptide ArgiLeu”’ and nested peptide Pro’-Arg” indicated that the NH2terminal boundary of this site is close to Arg2’. The greater potency of peptide Ala1s-Pro38 (IC5o = 86.6 X 10m6 M) as compared to Ala”-Leu”” (ICso = 119 x 10mh M) indicated that the four amino acids Thr”5-Tyr3h-Leu37-Pro3H are important for interactions with hormone. The potency of peptides Leu2y-Arg4n and Arg”‘-Arg4’, and the inactivity of peptide boundary of Pro4”-Asp “, indicated that the COOH-terminal this hormone-binding domain is between I’ro3H and Va14’. Va14’ was chosen as a possible COOH-terminal boundary because the next 6 amino acids (Ile42-Pro43-Ser44-Gln4s-Ala46Phe4’) are conserved residues in a larger exon-dependent repetitive motif recently described by Koo et al. (8). Previous studies have shown that both the 01- and P-subunits of hCG bind to receptor (1, 34, 35) and that at least three discrete regions on the a-subunit (17, 18, 22) and two on the Psubunit (20, 36) are involved in receptor contact. We think it unlikely that any of the residues in a repetitive motif (i.e. Ile4’-Phe47) would interact with several different sites on the hormone. Therefore, nested peptides Arg2’-Pro3X and Arg2’-Va14’ were synthesized and tested. As shown in Table 1, these two peptides have a 6- to IO-fold greater potency as compared to other peptides in this site, and the slightly greater activity of Arg2’-Pro”X maximally delimits a binding determinant within these residues. The homologous region of the human TSH receptor has also been demonstrated to be important in hormone binding (37, 38). The TSH receptor has an eightamino acid insertion, not present in the LH/hCG receptor, that immediately precedes the TSH receptor residue corresponding to LHR Arg*‘. Deletion or substitution of this eightamino acid segment by site-directed mutagenesis abolishes high affinity TSH binding (37). Our peptide data provides evidence that this region of the LH/hCG receptor is also important for interaction with hormone. Alignment of LHR residues 21-38 with the corresponding sequences in the TSH

Dose-Response 97.00

RECEPTOR

r

Curves for LHR Peptides r

r

, 97.0

95.0 90.0

FIG. 2. Competitive inhibition of specific T-hCG binding to rat luteal membranes by LHR peptides. ““I-hCG (2 ng) was incubated with 1.25 mg (wet weight) of 2000 x g luteal membrane in the presence or absence of varying doses of LHR peptides for 16 h at 22 C.

80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 7.0 IO'

102

103

10-d

IO'

Concentration

10"

103

IO-'

(uM)

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.

mo 2 8

212

HORMONE-BINDING SEQUENCE

COMPARISON

REGIONS OF

OF

ACTIVE

THE

LH/hCG

rLH/hCG

Endo. Voll31. No

RECEPTOR

RECEPTOR

1

REGIONS Identity

to

rLH/hCGr

21-38 Exon 1 Arg-Cys-Pro-Gly-Pro-Arg-Ala-Gly-Leu-Ala-Arg-LeIu-Ser-Leu-Thr-Tyr-Leu-Pro Arg-CqIs-Pro-Gly-Pro-Thr-Ala-Gly-Leu-Thr-Arg-LeIu-Ser-Leu-Ala-Tyr-Leu-Pro

Exon

rLH/hCGr: hLHr: hTSHr: rFSHr:

Arg-Ile-Pro-Ser-Leu-Pro-Pro-Ser-Thr-Gln-Thr-Lelu-Lye-Leu-Ile-Glu-Thr-H~e Pro-Thr-Asp-Leu-Pro-Arg-Aen-Ala-Ile-Glul **-Lelu-Arg-Phe-Val-***-Leu-Thr

rLH/hCGr: hLHr: hTSHr: rFSHr:

Exon 4 Arg-Leu-Lys-Tyr-Lelu-Ser-Ile-Cys-Asn-Thr-Gly-Ile-Arg-Thr Arg-Leu-Lys-Tyr-LeIu-Ser-Ile-Cys-Asn-Thr-Gly-Ile-Arg-Lys Leu-Leu-Lys-Phe-Lelu-Gly-Ile-Phe-Asn-Thr-Gly-Leu-Lya-Met Ser-Leu-Arg-Tyr-LeIu-Ser-Ile-Ser-Asn-Thr-Gly-Ile-Lye-Hle

rLH/hCGr: hLHr: hTSHr: rFSHr:

253-212 Exon 9 ~r-Pro-Ser-His-Cys-Cys-Ala-Phe-Arg-Asn-Leu-Pro-Lys-Lys-Gllu-Gln-Asn-Phe-Ser-Phe Tyr-Pro-Ser-His-Cys-Cys-Ala-Phe-Arg-Asn-Leu-Pro-Thr-Lys-GlIu-Gln-Asn-Phe-Ser-His Tyr-Pro-Ser-His-Cys-Cys-Ala-Phe-Lys-Asn-Gln-Lys-Lys-Ile-ArIg-Gly-Ile-Leu-Glu-Ser Tyr-Pro-Ser-His-Cys-Cys-Ala-Phe-Ala-Asn-Leu-Lys-Arg-Gln-IlIe-Ser-Glu-Leu-His-Pro

rLH/hCGr: hLHr: hTSHr: rFSHr:

Exon 11 Lys-Val-Pro-Leu-Ile-Thr-Val-Thr-Asn-Ser-Lys Lys-Val-Pro-Leu-Ile-Thr-Val-Thr-Asn-Ser-Lys Asn-Lys-Pro-Leu-Ile-Thr-Val-Ser-Asn-Ser-Lys Lys-Val-Pro-Leu-Ile-Thr-Val-Ser-Lys-Ala-Lys

2 100% 83% 22% 22%

102-115 Exon

5 100% 93% 50% 64%

Exon

10

100% 90% 50% 50%

573-583 100% 100% 73% 73%

FIG. 3. Sequence alignment of active rat LH/hCG Ref. 14), the human TSH receptor (hTSHr, Ref. sequences are shown in boldface type. uertical bars denote exon-exon junctions

receptor (rLH/hCGr) regions with homologous sequences of the human LH receptor (hLHr, 7), and the rat FSH receptor (rFSHr, Ref. 5). Residues that are not identical to rLH/hCGr The triple asterisk represents residue deletions at the corresponding position to the rLH/hCGr sequence; as described by Kookt al. (8).

(7) and FSH (5) receptors (Fig. 3) indicated that this is a region of relatively low homology (22% identity), and therefore may constitute a site that interacts with the P-subunit of hormone. A second binding region was initially identified with peptide Asn”-Asp”‘. Overlap peptides on both the NH*-terminus (AsnK5-Lys”‘4) and COOH-terminus (Ile”“-Glu’32) were without activity in radioreceptor assays, suggesting that the central residues in the peptide (Tyr’““-Gly”*) are essential for activity. We therefore synthesized the nested peptide Arg’02-Thr”5 and found that it had increased potency (Table 1). Sequence alignment and comparison of Arg’02-Thr’15 with the other glycoprotein hormones receptors (Fig. 3) reveals a strong degree of homology (~50% identity), and suggests that this segment may bind the a-subunit of hormone. Overlap peptides Ser23y-Cys258and Tyr253-Phe272defined a third hormone-binding site. The more than 3-fold greater potency of peptide Tyr253-Phe272 over peptide Ser23y-Cys258 suggested that the binding site was contained within the overlap residues (Tyr253-CYSTS*)and core residues Ala’“‘Lys2h6.Because the peptide overlapping the COOH-terminus of Tyr253-l?he272(G1uzh7-ValzR6)had no activity, we hypothesized that LYS*~‘~defined the COOH-terminal end of this binding site. We therefore synthesized the nested peptide Tyr253-Lys*‘j’. To our surprise, it had nearly 3-fold lower potency than Tyr253-Phe272,indicating that several, if not all,

of the residues G1u2h7-Gln2h8-Asn26y-Phe270-Ser27~-Phe272 participate in hormone binding. Sequence alignment of LHR residues of Tyr253-Phe272with residues of the human TSH receptor and rat FSH receptor revealed a high degree of amino acid similarity among the receptors (50% identity), indicating that these residues may constitute another contact site for hormonal a-subunit. A peptide replicating the third putative extracellular loop (3, 4), Lys573-LYs~~~(Y),defined a fourth receptor binding domain. Peptides replicating the three proposed extracellular loops were studied because of the high sequence homology for these regions in all glycoprotein hormone receptors (3-8, 33) and consequently their candidacy as potential a-subunit binding sites. Previous work by our laboratory has identified three effecter/receptor binding domains in the a-subunit of LH/hCG (17, 22). Based upon their activity and sequence homologies with the other glycoprotein hormone receptors, we hypothesize that three regions in the LH/hCG receptor, Arg’02-Thr’15, Tyr253-Phe272, and Lys573-Tyr583interact with n-subunit, although not necessarily in a cooperative manner as discussed below. Recent reports by Ji and Ji (13, 39) have described mutant LH/hCG receptors with truncated NH*-terminal domains. One of the mutants was composed only of exon 1 (Arg’Leu3*) and exon 11 (Tyr295-Hish74)whereas others contained essentially no NH*-terminal, extracellular extension. Surpris-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.

HORMONE-BINDING

REGIONS

ingly, all mutant receptors were capable of specifically bindalbeit with low affinity, and of stimulating ing ““I-hCG, CAMP production. In agreement with our findings, Ji and coworkers concluded that multiple contact points exist between the receptor and hormone, and that there is a hormone contact site in the COOH-terminal half of the receptor. We have identified two potential hormone-binding regions within exons 1 and 11, and have hypothesized that Arg”Pro”” is a P-subunit contact site, and that L~s’~~-Tyr”*” is an a-subunit binding site. Several reports have suggested that the a-subunit of hCG is in more intimate contact with the receptor when hormone is bound, and that the o-subunit is exposed on the surface of the hormone-receptor complex (13,40-43). A possible explanation, supported by our peptide data, is that there are multiple (three) contact sites for Qsubunit and only one contact site for @-subunit. The specific but low affinity binding of the exon 1 and exon 11 mutant receptor may therefore be due to the presence of a P-subunit contact site and only one a-contact site. Conversely, maintenance of high affinity binding with a mutant receptor composed of exons l-10 (13) may be due to the presence of one P- and two (Arg”‘2 -Thr’lS and Tyr253-Phe272) of the three a-subunit contact regions. Nagayama et al. (38) have constructed chimeric TSH-LH/hCG receptors and have described one that bound both TSH and hCG. This particular chimera was composed of the TSH receptor with the first 170 amino acids substituted by the corresponding NHzterminus of the LH/hCG receptor. Our findings would suggest that the activity of such a chimera is due to the presence of an LH/hCG P-subunit binding region (Arg2’-I’ro3H) and an a-subunit binding region (Arg’“2-Thr”5). The potency of each synthetic peptide, expressed as the molar concentration causing half-maximal inhibition of binding of 12’1-hCG, is summarized in Table 1. The low affinity of peptides has been a recurrent criticism of the synthetic peptide approach for identifying specific binding/effecter sites. The affinity of active peptides (Kd = lo-“ to 10m5 M) compared to intact receptor (Kd = lo-‘” M) is reasonable and is most likely due to multiple (as our data supports), discontinuous sites in the native receptor that interact in a cooperative manner with hormone. Specific binding of this order of magnitude has been used to define receptor binding domains in the a-subunit of the glycoprotein hormones (1719, 22) and in the /?-subunits of LH/hCG (20, 36), TSH (21), and FSH (23, 24). Other laboratories have also used synthetic peptides in competitive binding assays to define ligandreceptor interactions involving coagulation factors (26), lymphocyte (27), and platelet adhesion (28), and cholinergic binding sites (29, 30). In addition, Atassi et al. (25) have recently reported on the use of synthetic peptides to identify hormone binding regions in the TSH receptor but their peptides displayed surprisingly higher affinities (Kd > lo-’ M), perhaps due to methodological differences in assessing binding. The fact that the majority of peptides tested had no inhibitory activity (even those with overlapping residues) readily indicates the specificity of this approach. If our observed inhibitory activity were nonspecific, the nearly 9-fold greater potency of peptide Arg2’-Pro38 over peptide Ala15-

OF THE

LH/hCG

RECEPTOR

273

Pro3X is very difficult to explain. The extracellular domain of the rat LH/hCG receptor has six potential sites for N-linked glycosylation (3). Of the four hormone-binding regions that we have identified only one (Tyr2s3-Phe272) contains a potential glycosylation site (Asn2hy). As discussed above, the COOH-terminal residues of this region appear to be important in hormone binding. Previous works of several investigators (44, 45) have shown that the receptor N-oligosaccharides are not required for high affinity binding of hormone and have suggested that the carbohydrate moieties are removed from hormone contact regions. Proof of a carbohydrate chain on Asn26y and its possible influence on hormone binding awaits further study. N-Glycosylation may, however, be necessary for proper folding and functional expression of glycoprotein hormone receptors on the cell surface (45, 46). Continued characterization of the structural and functional properties of the LH/hCG receptor is essential in order to understand the molecular events that occur during the interaction of gonadotropins, LH/hCG receptors, and intracellular effector systems. We have demonstrated in this report that synthetic peptides replicating specific receptor regions can be used to identify hormone interaction sites. These sites can now be targeted for modification by site-directed mutagenesis to further define the essential contact amino acids. Evaluation of all peptides for their influence on adenylyl cyclase activity is in progress. Acknowledgments We are grateful to B. J. Madden and M. C. Charlesworth for their assistance in the synthesis and sequencing of the peptides used in these studies, and to Kathleen Kitzmann for excellent technical assistance.

References 1 Roche PC, Ryan RJ 1985 The LH/hCG receptor. In: Ascoli M (ed) Luteinizing Hormone Action and Receptors. CRC Press, Boca Raton, FL, DD 17-56 2. Hunzicker-Dunn M, Birnbaumer L 1985 The involvement of adenylyl cyclase and cyclic AMP-dependent protein kinases in luteinizing hormone action. In: Ascoli M (ed) Luteinizing Hormone Action and Receptors. CRC Press, Boca Raton, FL, pp 57-134 KC, Sprengel R, Phillips HS, Kohler M, Rosemblit N, 3. McFarland Nikolics K, Segaloff DL, Seeburg PH 1989 Lutropin-choriogonadotropin receptor: an unusual member of the G protein-coupled receptor family. Science 245:494-499 4. Loosfelt H, Misrahi M, Atger M, Saleasc R, Thi MTVH-L, Jolivet A, Guiochon-Mantel A, Sar S, Jallal B, Garnier J, Milgrom E 1989 Cloning and sequencing of porcine LH-hCG receptor cDNA: variants lacking transmembrane domain. Science 245:525-528 5. Sprengel R, Braun T, Nikolics K, Segaloff DL, Seeburg PH 1990 The testicular receptor for follicle stimulating hormone: structure and functional expression of cloned cDNA. Mol Endocrinol 4:525530 6. Parmentier M, Libert F, Maenhaut C, Lefort A, Gerard C, Perret J, Van Sandio J, Dumont JE, Vassart G 1989 Molecular cloning of the thyrotropin receptor. Science 246:1620-1622 Misrahi M, Loosfelt H, Atger M, Sar S, Guiochon-Mantel A, Milgrom E 1990 Cloning, sequencing, and expression of human TSH receptor. Biochem Biophys Res Commun 166:394-403 Koo YB, Ji I, Slaughter RG, Ji TH 1991 Structure of the luteinizing hormone receptor gene and multiple exons of the coding sequence. Endocrinology 128:2297-2308 Tsai-Morris CH, Buczko E, Wang W, Xie X-Z, Dufau ML 1991 Structural organization of the rat luteinizing hormone (LH) receptor

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.

274

HORMONE-BINDING gene. J Biol Chem

10.

266:11355-11359 domain of receptor expressed in transfected cells with high affinity. J Biol Chem

Tsai-Morris

Moyle WR, Bernard

hormone hormone

receptor binding

gene defines a soluble activity. J Biol Chem

MI’, Myers RV, Marko

&I.

28.

Endo. Voll31.

RECEPTOR which

contains

1992 No 1

the CSl

DSouza SE, Ginsberg MH, Burke TA, Plow EF 1990 The ligand binding site of the platelet integrin receptor GPIIb-IIIa is proximal to the second calcium binding domain of its alpha subunit. J Biol Chem 265:3440-3446

29.

Conti-Tronconi hardt-Maelicke

BM, Tang F, Diethelm BM, Spencer SR, ReinS, Maelicke A 1990 Mapping of a cholinergic

binding site by means and alpha-bungarotoxin.

OM, Strader CD

1991 Leutropin/beta-adrenergic receptor chimeras bind choriogonadotropin and adrenergic ligands but are not expressed at the cell surface. J Biol Chem 266:10807-10812 13. Ii I. Ii TH 1991 Exons I-10 of the rat LH receptor encode a high affimty hormone binding site and exon 11 encodes G-protein modulation and a potential second hormone binding site. Endocrinology 128:2648-2650 14. Jia X-C, Oikawa M, Bo M, Tanaka T, Ny T, Boime I, Hsueh AJW 1991 Expression of human luteinizing hormone (LH) receptor: interaction with LH and chorionic gonadotropin from human but not equine, rat, and ovine species. Mol Endocrinol 5:759-768 15. Lopez JA, Chung DW, Fujikawa K, Hagen FS, Papayannopoulou T, Roth GJ 1987 Cloning of the alpha chain of human platelet glycoprotein lb: a transmembrane protein with homology to leutine-rich alpha-2-glycoprotein. Proc Nat1 Acad Sci USA 84:56155619 16. Tan F, Weerasinghe DK, Skidgel RA, Tamei H, Kaul RK, Roninson IB, Schilling JW, Erdos EG 1990 The deduced protein sequence of the human carboxypeptidase N high molecular weight subunit reveals the presence of leucine-rich tandem repeats. J Biol Chem 265:13-19 17. Charlesworth MC, McCormick DJ, Madden BJ, Ryan RJ 1987 Inhibition of human choriotropin binding to receptor by human choriotropin (Y peptides. J Biol Chem 262:13409-13416 MC, McCormick DJ, Ryan 18. Morris JC, Jiang N-S, Charlesworth RJ 1988 The effects of synthetic e-subunit peptides on thyrotropin interaction with its receptor. Endocrinology 123:456-462 MC, McCormick DJ, 19. Morris JC, Jiang N, Hay ID, Charlesworth Ryan RJ 1988 The effects of synthetic alpha-subunit peptides on thyroid-stimulating immunoglobulin activity. J Clin Endocrinol Metab 67:707-712 MC, Mason KA, Ostrea T, Johnson 20. Keutmann HT, Charlesworth L, Ryan RJ 1987 A receptor-binding region in human choriogonadotropin/lutropin p subunit, Proc Nat1 Acad Sci USA 84:2038-2042 21. Morris JC, McCormick DJ, Ryan RJ 1990 Inhibition of thyrotropin binding to receptor by synthetic human thyrotropin fi-peptides. J Biol Chem 256:1881-1884 22. Erickson LE, Rizza SA, Bergert ER, Charlesworth MC, McCormick DJ, Ryan RJ 1990 Synthetic a-subunit peptides stimulate testosterone production in vitro by rat leydig cells. Endocrinology 126:2555-2560 23 Santa Coloma TA, Dattatreyamurty B, Reichert LE 1990 A synthetic peptide corresponding to human FSH P-subunit 33-53 binds to FSH receptor, stimulates basal estradiol biosynthesis, and is a partial antagonist of FSH. Biochemistry 29:1194-1200 24 Santa Coloma TA, Reichert Jr LE 1990 Identification of a folliclestimulating hormone receptor-binding region in hFSH-P-(81-95) using synthetic peptides. J Biol Chem 265:5037-5042 25 Atassi MZ, Manshouri T, Salata S 1991 Localization and synthesis of the hormone-binding regions of the human thyrotropin receptor. Proc Nat1 Acad Sci USA 88:3613-3617 26 Altieri DC, Etingin OR, Fair DS, Brunck TE, Geltosky JE, Haffar DP, Edgington TS 1991 Structurally homologous ligand binding of integrin Mac-l and viral glycoprotein C receptors. Science 254:1200-1202 77 Ager A, Humphries MJ 1990 Use of synthetic peptides to probe lymphocyte-high endothelial cell interactions. Lymphocytes recog-

LH/hCG

nize a ligand on the endothelial surface adhesion motif. Int Immunol 2:921-928

CH, Buczko E, Wang W, Dufau ML 1990 Intronic

nature of the rat luteinizing receptor subspecies with 265:19385-19388 12.

OF THE

Xie Y-B, Wang H, Segaloff DL 1990 Extracellular lutropin/choriogonadotropin binds choriogonadotropin 265:21411-21414

11.

REGIONS

30.

Maelicke A, Schroder B, Reinhardt-Maelicke elm BM, Conti-Tronconi BM 1991 Nicotinic have ligand-specific 435

31.

of synthetic peptides, monoclonal Biochemistry 29:6221-6230

attachment

point

antibodies,

S, McLane K, Dieth-

acetylcholine receptors patterns. J Recept Res 11:425-

Lee CY, Ryan RJ 1973 Interaction

of ovarian receptors with human luteinizing hormone and human chorionic gonadotropin. Biochemistry 12:4609-4615 32. Group GC 1984 A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12:387-395 33. Frazier AL, Robbins LS, Stork PJ, Sprengel R, Segaloff DL, Cone RD 1990 Isolation of TSH and LH/hCG receptor cDNAs from human thyroid: regulation of tissue specific splicing. Mol Endocrinol 4~1264-1276 34. Roche PC, Ryan RJ 1989 Purification, characterization, and aminoterminal sequence of rat ovarian receptor for luteinizing hormone/ human choriogonadotropin. J Biol Chem 264:4636-4641 35. Ascoli M, Segaloff DL 1986 Effects of collagenase on the structure of the lutropin/choriogonadotropin receptor. J Biol Chem 261:38073815 36. Keutmann HT, Mason LA, Kitzmann KA, Ryan RJ 1989 Role of the &93-100) “determinant loop” sequence in the biological activity of human lutropin and choriogonadotropin. Mol Endocrinol 3:526533 37. Wadsworth HL, Chazenbalk GD, Nagayama T, Russo D, Rapoport B 1990 An insertion in the human thyrotropin receptor critical for high affinity hormone binding. Science 249:1423-1425 38. Nagayama Y, Wadsworth HL, Chazenbalk GD, Russo D, Seto I’, Rapoport B 1991 Thyrotropin-luteinizing hormone/chorionic gonadotropin receptor extracellular domain chimeras as probes for thyrotropin recehtor function. Proc Nat1 Acad Sci USA 88:902-905 39. Ji I, Ji TH 1991 Human choriogonadotropin binds to a lutropin receptor with essentially no N-terminal extension and stimulates CAMP synthesis, J Biol Chem 266:13076-13079 40. Milius RI’, Midgley Jr AR, Birken S 1983 Preferential masking by the receptor of immunoreactive sites on the DI subunit of human choriogonadotropin. Proc Nat1 Acad Sci USA 80:7375-7379 41. Movle , WR, Ehrlich PH. Canfield RE 1982 Use of monoclonal antibodies to subunits of human chorionic gonadotropin to examine the orientation of the hormone in its complex with receptor. Proc Nat1 Acad Sci USA 79:2245-2249 42. Kusuda S, Dufau ML 1986 Purification and characterization of the rat ovarian receptor for luteinizing hormone. J Biol Chem 261:16161-16168 43. Petiji U, Kellokumpu S, Keinanen K, Metsikko K, Rajaniemi H 1987 Covalent cross-linking of radiolabeled human chorionic gonadotropin to rat ovarian luteinizing hormone receptor with glutaraldehyde. J Recept Res 7:809-827 44. Keinanen K 1988 Effect of deglycosylation on the structure and hormone-binding activity of the lutropin receptor. Biochem J 256:719-724 45. Ji I, Slaughter RG, Ji TH 1990 N-linked oligosaccharides are not required for hormone binding of the lutropin receptor in a Leydig tumor cell line and rat granulosa cells. Endocrinology 127:494-496 46. Russo D, Chazenbalk GD, Nagayama Y, Wadsworth HL, Rapoport B 1991 Site-directed mutagenesis of the human thyrotropin receptor: role of asparagine-linked oligosaccharides in the expression of a functional receptor. Mol Endocrinol5:29-33

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 07:11 For personal use only. No other uses without permission. . All rights reserved.