Molecular and Cellular Endocrinology, 86 (1992) 37-47 0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/92/$05.00

MOLCEL

37

02777

Functional expression of an ovine growth hormone receptor in transfected Chinese hamster ovary cells Rodney J. Fiddes, Malcolm R. Brandon and Timothy E. Adams Centre

for Animal Biotechnology, School of Veterinary Science, The Unil,ersity of Melbourne, Park{Y//e, Vtctoria 3052, Australia (Received

Key words: Growth

hormone

receptor;

Functional

30 December

expression;

1991; accepted

CHO

cell;

13 March

1992)

Ovine placental lactogen: (Sheep)

Summary Structural heterogeneity has been demonstrated for growth hormone (GH) receptors from a number of species, and both high and low affinity art receptors have been characterised by ligand binding studies. In the present study, we have transfected Chinese hamster ovary (CHO-Kl) cells with a cDNA clone encoding a full-length transmembrane ovine (01 GH receptor, under the regulatory control of the human metallothionein IIA promoter. A stably transfected cell line was established (GHR9.5) which expresses on the cell surface a single class of receptor which binds 220,000 [‘2”I]oGH molecules at high affinity (Kd = 0.30 nM) which is comparable to the affinity established for endogenous oGH receptors in postnatal sheep liver microsomes (Kd = 0.27 nM, Freemark et al. (1987) Endocrinology 120, 1865-1872). The expressed receptor also binds ovine placental lactogen (oPL, 205,000 binding sites per cell) with high affinity (Kd = 0.76 nM). The presence of two species of oGH receptor was detected in GHR9.5 cells using affinity cross-linking analysis (M, 148,000 and M, 73,000) and given that the oGH receptor cDNA codes for a non-glycosylated receptor of M, 69,914, it is likely that these cross-linked species correspond to homodimeric and monomeric forms of the oGH receptor, each binding to a single molecule of GH. Parallel cross-linking studies with sheep liver microsomes also demonstrated two oGH receptor species (M, 133,000 and M, SS,OOO>,the difference in relative molecular weights between the transfected and endogenous receptors presumably resulting from tissue-specific post-translational modifications. In the presence of oGH, the GHR9.5 cells respond by increasing total cellular protein synthesis by 27% relative to non-GH-exposed GHR9.5 cells, indicating the functionality of the expressed receptor. We also demonstrate unequivocally that oPL, through a specific interaction with the transfected oGH receptor, is able to mediate a similar cellular response (38% protein synthesis induction). Responsiveness to oGH and oPL in the GHR9.5 cells is dependent on serum starvation prior to oGH exposure and occurs only with prolonged exposure (greater than 2 h) to oGH. This cellular stimulation occurs independently of c-fos transcription which has previously been shown to be one of the earliest events associated with GH action in tissues expressing endogenous GH receptors (Doglio et al. (1989) Proc. Natl. Acad. Sci. USA 86. 1148-1152; Slootweg et al. (1990) J. Mol. Endocrinol. 4, 265-274).

Introduction Correspondence to: Dr. Tim Adams, Centre for Animal Biotechnology, School of Veterinary Science, The University of Melbourne, Parkville, Victoria 3052, Australia. Tel. (03) 344-7362; Fax (03) 347-4083.

The isolation of cDNA clones encoding growth hormone (GH) receptors from a number of

species has provided, in part, an explanation for the biochemical heterogeneity originally reported for this receptor (Barnard et al., 1985; Ymer and Herington, 1985; Baumann et al., 1986; Herington et al., 1986). Characteristically, cDNA clones encoding a full-length GH receptor predict a membrane-bound protein of 614-626 amino acids, comprising a ligand-binding extracellular domain, a hydrophobic transmembrane region and an extended cytoplasmic domain (reviewed by Mathews, 1991). While only a single GH receptor gene has been identified in any one species (Godowski et al., 1989) alternate splicing generates multiple GH receptor mRNA transcripts which may code for truncated (Leung et al., 1987) or soluble (Smith et al., 1988; Baumbach et al., 1989) GH receptors in a number of species. Interestingly, multiple mRNA transcripts code for structurally diverse species of the prolactin (PRL) receptor (Boutin et al., 1989; Davis and Linzer, 1989), a polypeptide receptor with which the GH receptor shares sequence similarity. The mechanism(s) by which the GH receptor effects cellular responses as a result of GH binding remain largely undefined. Phosphorylation of the GH receptor on tyrosine residues following ligand binding has been reported (Foster et al., 19881, which appears mediated by a GH receptor-associated kinase (Carter-Su et al., 1989). Significantly, the GH receptor is not itself a tyrosine kinase. GH can also induce changes in G-protein function in adipose plasma membranes (Roupas et al., 1991) and stimulates c-fos mRNA expression independent of phosphoinositide or inositol lipid turnover (Doglio et al., 1989; Slootweg et al., 1990). Despite the structural heterogeneity associated with the GH receptor, as a result of alternate mRNA splicing or extensive post-translational modification of membranebound GH receptors during their biosynthesis (Hocquette et al.. 1990; Moldrup et al., 19901, at least one study supports the hypothesis that the full-length GH receptor is a prerequisite for signal transduction following GH binding. Transfection of a cDNA clone encoding the full-length rat GH receptor into the rat insulinoma cell line RIN-5AH results in transfectant cell lines expressing elevated levels of rat GH receptors relative to the insulinoma parent cell line, and significantly augments GH stimulation of insulin

biosynthesis (Billestrup et al., 1990). In contrast, cell lines expressing COOH-terminal truncated, membrane-bound GH receptor variants do not exhibit enhanced insulin biosynthesis in response to GH (Moldrup et al., 1091). The versatility of cellular transfection as an assay for GH receptor function is emphasised by the work of Emtner et al. (1990) who have shown that expression of rat GH receptors by transfected Chinese hamster ovary (CHO) cells can mediate the stimulation of protein synthesis in these ceils following GH binding. Despite elevated levels of circulating GH in fetal serum, the stimulation of somatic growth by GH in sheep is observed only in postnatal animals. This results from the developmental regulation of ovine to) GH receptors; few, if any, hepatic GH receptors are found in fetal sheep prior to parturition, following which they are readily detected on hepatic membranes within the first week of postnatal development (Gluckman et al., 1983; Freemark et al., 1986). Similarly, the appearance of oGH receptor mRNA transcripts in the fetal liver is not found until late gestation (Adams et al., 1990). Heterogeneity has been described for oGH receptors expressed on liver membranes, with both high-affinity and low-affinity species of receptor being detected (Sauerwein et al., 1991). Although the structural relationship between these receptors has not been established, there is evidence to suggest they are independently regulated (Sauerwein et al., 1991). We have previously isolated cDNA clones coding for a full-length, transmembrane oGH receptor (Adams et al., 1990). In the present study we have established a transfected cell line overexpressing oGH receptors in order to examine the homogeneity of the cell surface receptor by ligand-binding and cross-linking studies. Furthermore, we have examined the ability of transfected oGH receptors to mediate signal transduction following binding of not only GH, but also in the presence of the related somatogenic polypeptide hormone, ovine placental lactogen (oPL). Materials

and methods

Materials Recombinant was purchased

bovine growth from Monsanto

hormone (bGH) (USA), recombi-

nant oPL was a gift from Professor Peter Gluckman (University of Auckland, New Zealand) and native oGH was obtained from the National Institute of Health Hormone Distribution Program (NIADDK-oGH-I-4 for iodinations and NIADDK-oGH-15). Restriction and DNA modifying enzymes were obtained from Pharmacia LKB Biotechnology (Melbourne, Australia) and Boehringer-Mannheim (Melbourne, Australia). All chemicals used were of molecular biology or analytical grade. Iodina tions Hormones (bGH, oPL, oGH) were labeled using the catalyst Iodogen (Pierce, USA). Briefly, 10 pg of hormone was incubated with 1 pg of Iodogen plus 40 I_LIof 50 mM sodium phosphate, pH 7.4 and 1 mCi Na “‘1 (IMS-30, 100 mCi/ml, Amersham, Australia) for 10 min at room temperature. Free iodine was separated from labeled hormone by column chromatography (40 x 1 cm) using Sephadex G-100 (Pharmacia, Australia) and an elution buffer of 0.2% bovine serum albumin (BSA, fraction V, Boehringer-Mannheim) in 50 mM sodium phosphate, pH 7.4. Specific activity of labeled hormones was 40-80 pCi/pg. Expression of the ocine GH receptor A full-length oGH receptor cDNA clone was assembled from two overlapping clones obtained from a sheep liver library (Adams et al., 1990, GenBank accession number M82912). Briefly, a 1082 bp DraI/EcoRI fragment from the hsGHR2 clone and a 996 bp EcoRI/DraI fragment from the AsGHRl clone were sequentially subcloned into the eukaryotic expression vector FIII (Mountford, 1991). The FIII vector is a derivative of the plasmid SP72 (Promega, Australia) and contains the human metallothionein (MTIIA) promoter in combination with an SV40 enhancer (Friedman et al., 1989). This vector utilises the SV40 small t antigen intron, and polyadenylation and transcriptional termination signals originally from the plasmid pkc-neo (Van Doren et al., 1984). All cloning techniques were performed according to standard procedures (Sambrook et al., 1989). Chinese hamster ovary (CHO-KI, American Type Tissue Collection) cells were maintained in

a-modified Eagle’s medium (a-MEM, Gibco, Detroit, USA) plus 10% fetal calf serum (FCS) at 37°C in a humidified atmosphere of 5% CO,. Subconfluent cultures were transfected by calcium phosphate co-precipitation (Wigler et al., 1979) with 10 pg of GHR-FIII and 0.5 Kg of the selection plasmid pSV2.gpt (Mulligan and Berg, 1981). Transfected cells were then glycerol shocked (15% glycerol for 3 min at 37°C) and following overnight incubation in a-MEM medium plus 10% FCS, cells were grown in selection medium (cu-MEM/lO% FCS supplemented with 50 U/ml penicillin, 50 kg/ml streptomycin, 250 pg/ml xanthine, 10 pg/ml mycophenolic acid, 0.1 mM hypoxanthine, 0.4 PM aminopterin, 16 PM thymidine) for 7-10 days. Stable transfectants were then isolated by dilution cloning and expanded in 10 cm2 tissue culture dishes. Clones expressing the membrane-bound receptor were detected using a hormone binding assay. Hormone binding assays Confluent cultures of stable transfectants were washed in serum-free medium and incubated in the same medium for 2 h at 37°C to minimise the influence of serum-derived GH on the binding assay. After the medium was aspirated, cells were incubated for 4 h at room temperature (- 2l’C) in 5 ml 0.5% BSA/phosphate buffered saline, pH 7.4 (PBS) containing 4 X 10’ cpm of [12’I]bGH. Following three washes in 10 ml cold (4°C) PBS, cells were solubihsed in 0.1 M NaOH and the bound radioactivity quantified in a gamma counter (Packard 5000. Scatchard plots Confluent cultures of GHR9.5 cells were serum-starved for 2 h and collected in 5 mM ethylenediaminetetraacetic acid tEDTA)/PBS and after centrifugation were resuspended in binding assay buffer (50 mM Tris-HCl pH 7.5, 10 mM MgCl,, 0.1% BSA, 0.02% benzamidine-HCI). Varying levels of [12”I]bGH or [ ‘*‘I]oPL were added to duplicate assay tubes, containing GHR9.5 cells (1.0 X 10”) in a total volume of 400 ~1 of binding assay buffer. After overnight (18 h) incubation at 4°C 3 ml of cold assay buffer was added, the tubes centrifuged (10 min at 750 x g) and the radioactivity bound in the cell pellet

40

quantified. Non-specific binding was evaluated for each assay point by the addition, to identical duplicate tubes, of 1 pg of cold oGH (NIADDKoGH-15). The amount of radiolabeled hormone bound to the cells, the amount of hormone not bound (free) and the ratio of bound to free hormone were evaluated for each assay point and presented as Scatchard plots (Scatchard, 1953).

Protein synthesis The ability of oGH and oPL to stimulate protein synthesis in GHR9.5 cells was measured by the incorporation of [ “Hlleucine (~-[4,5-~Hlleucine, 47 Ci/mmol, Amersham, Australia) into trichloroacetic acid (TCA) precipitable proteins, essentially as described by Emtner et al. (1990). Briefly, duplicate confluent cultures (6 cm dishes) were serum starved for 24 h, incubated with hormone (1 pg/ml oGH or oPL) in serum-free growth medium for the indicated time periods (see Results), and pulse labeled with 3 PCi [‘Hlleucine for the final 2 h of incubation. After the cells were solubilised in 0.1% sodium dodecyl sulphate (SDS), TCA was added to 8% (v/v) and precipitable proteins collected on Whatman GF/ C filters (Whatman International, Maidston, UK). Radioactivity was quantified by liquid-scintillation counting.

Northern blots Confluent cultures of GHR9.5 cells were serum-starved for 24 h and then exposed to either growth medium plus FCS alone, growth medium plus FCS plus 1 pg/ml oGH or growth medium plus 1 Kg/ml oGH alone. After 15 min incubation, total RNA was isolated from these cells by acid-phenol guanidine thiocyanate extraction (Chomczynski and Sacchi, 1987) and RNA samples fractionated by electrophoresis through denaturing formaldehyde agarose gels and transferred to Hybond Nt nylon membranes (Amersham). The membranes were hybridised to j2Plabeled cDNA probes generated by random priming, essentially as described in Sambrook et al. (1989).

Cross-linking Liver microsomes were purified using differential centrifugation (Ghan et al., 1978) and endoge-

nous GH stripped by treatment with MgCl? (Baxter et al., 1984). GHR9.5 cells were serum starved for 2 h to minimise the binding of the expressed GH receptor with the bGH which is contained in the FCS portion of the complete medium. GHR9.5 cells were then collected in 5 mM EDTA in PBS. Liver microsomes (100 pg protein) or GHR9.5 ceils (1 x 106) were incubated with labeled hormone (3 x 10’ cpm), for 4 h at room temperature, in 500 ~1 binding assay buffer. After two washes in 10 ml PBS, GHR9.5 cells or microsomes were resuspended in 50 mM sodium phosphate, pH 7.4 containing 0.5 mM disuccininidyl suberate (DSS, Pierce, USA) and incubated for 15 min at room temperature. After washing in 10 ml 10 mM Tris-HCI pH 7.4, 1 mM EDTA, the GHR9.5 cells or microsomes were solubilised, in the absence or presence of 2% /3-mercaptoethanol, in 100 ~1 of electrophoresis sample buffer (0.5 M Tris pH 6.8, 5% SDS, 20% glycerol, 0.005% bromophenol blue) then boiled and electrophoresed on SDS-acrylamide (8%) gels, according to the method of Laemmli (1970); gels were then dried and autoradiographed against Fuji RX X-ray film at -70°C. Results

Isolation of stable transfectants expressing the ol’ine GH receptor A cDNA insert encoding a putative full-length oGH receptor was assembled from two cDNA clones isolated previously (Adams et al., 1990), and placed downstream of a human MTIIA promoter, linked to an SV40 enhancer (Fig. 1). The resulting plasmid, designated GHR : FBI, was initially tested for its ability to drive the expression of a functional cell-surface oGH receptor by transient expression studies in COS cells (Gluzman, 1981). Transfected cell cultures displayed significant levels of surface-bound [ ‘2s1]bGH in ligandbinding studies (data not shown) confirming the identity of the polypeptide product encoded by the isolated cDNA clones as the oGH receptor. Stable, transfected cell lines expressing the oGH receptor were established by co-transfecting CHO-Kl cells with GHR: FIB and the plasmid pSV2.gpt, encoding the dominant bacterial selection marker xanthine-guanine phosphoribosyl

41

transferase (Mulligan and Berg, 1981). Following selection, clonal cell lines were established and expanded for analysis in ligand-binding assays. Of ten cloned cell lines assayed, one bound GH at the same level as the control (non-transfected CHO cells), eight bound GH at approximately 2-fold greater than the level of the control and one, designated GHR9.5, bound GH at greater than lo-fold that of the control (data not shown). Clone GHR9.5 was utilised for the studies presented here. Scatchard analysis of ligand-receptor interaction The binding of radiolabeled ligands to oGH receptors expressed by GHR9.5 was analysed using conventional Scatchard plots (Scatchard, 1953). Recombinant bGH, overexpressed and purified from Escherichia co/i, was radioactively labeled and used in the binding assays. Bovine GH is almost identical to oGH at the level of amino acid homology (Warwick and Wallis, 1984) and with respect to biological activity (Bass et al., 1991; Sauerwein et al., 1991). For comparison, Scatchard analysis was also performed using bacterially-derived recombinant oPL, a placental hormone that binds to the hepatic oGH receptor

Smal/Dral Hindlll

,,~ ,,,,,, llllllllll’lll” 11@

hMTllA

E= e c [SV40E s g % Xhol

’

GHRlFlll 6.90 Kb 1

sv40tg-

Dral/Smal

‘T’/Hpal Fig. 1. Mammalian expression vector GHR: FlII encoding the oGH receptor. A DraI cDNA fragment, encoding the fulllength oGH receptor, was ligated into the FIII eukaryotic expression vector. The FIII vector contains the human metallothionein (hMTIIA) promoter in combination with an SV40 enhancer (SV40E) and utilises the SV40 small t antigen, and polyadenylation site and transcription termination signals (SV40t).

8

600

7 E

0

50

100

150

200

2io

molecules/cell Fig. 2. Scatchard analysis of binding of [ ‘*‘I]oGH CO) and [‘2sI]oPL (0) to GHR9.5 cells. Cells (I x 10’) were incubated with varying levels of labeled hormone. After overnight incubation and addition of 3 ml cold assay buffer, cells were collected by centrifugation and the counts bound quantified. The ratio of bound to free hormone was plotted against the amount of bound hormone and curves of best fit were evahated by simple regression analysis (Computer Associates, California, for Macintosh).

with high affinity (Freemark et al., 1986, 1987). The results of this analysis are presented in Fig. 2 and indicate that each GHR9.5 cell has an average of 220,000 binding sites for [‘*“I]bGH and 205,000 binding sites for [12sI]oPL and that oGH and oPL both show a high affinity for the expressed receptor (K,, = 0.30 nM and K, = 0.76 nM respectively). The plots obtained also demonstrate that for each ligand used, the data argues for a single class of receptor (as indicated by a straight line) being expressed on the GHR9.5 cells. Cellular responskeness of GHR9.5 cells to oGH and oPL The binding of GH to its receptor can initiate a variety of cellular responses, including the induction of protein synthesis (for review see Isaksson et al., 1985) and the transcriptional activation of the nuclear proto-oncogene c-fos (Doglio et al., 1989; Gurland et al., 1990; Slootweg et al., 1990). To analyse the functional status of the expressed oGH receptors on GHR9.5 cells in response to hormone ligands, duplicate cultures of confluent, serum-starved GHR9.5 cells were

32

entire 7 h incubation show an increase of 30% over control cultures. However, the addition of 10% FCS in a-MEM induced a rapid increase in protein synthesis of approximately 80% after only 2 h incubation (compared to no increase in protcin synthesis by oGH nor oPL in the first 2 h) in both GHR9.5 and non-transfected CHO cells cultures. In contrast, oGH failed to induce cTf0.r transcription in GHR9.S cells (Fig. 4). Northern blot analysis of total RNA isolated from serum-starved GHR9.S cells exposed to cu-MEM alone, u-MEM + oGH, a-MEM + FCS or (u-MEM + FCS + oGH for 15 min reveals similar levels of a 4.0 kb oGH receptor mRNA transcript in all cultures (Fig. 4A). Howcvcr, hybridising the same total RNA to a ‘*P-labeled human c-fos probe showed a message of 2.2 kb in total RNA from serumstarved cells incubated for IS min with either FCS or FCS + oGH, but not in non-stimulated nor oGH-treated cells (Fig. 48). Identical results were found when oPL was substituted for oGH (not shown). These results indicate that oGH and oPL do not stimulate the transcription of c;fos in

incubated with n-MEM medium containing pituitary oGH or recombinant oPL, and the induction of protein synthesis monitored by the incorporation of [‘Hlleucine into TCA-precipitable proteins. Both oGH and oPL stimulated significant increases in protein synthesis in GHR9.5 cells (Fig. 3A) with maximal stimulation being observed Y h after the addition of hormone. The level of induction observed at this time with oGH and oPL was 28% and 37% respectively, relative to GHRY.S cells exposed to (u-MEM alone (controls). Non-transfected CHO cells show the same levels of protein synthesis as the controls when incubated with oGH or’oPL. No further increase in protcin synthesis levels was detected beyond Y h of exposure to oGH or oPL in GHRY.5 cells (not shown). Stimulation of protein synthesis did not occur in serum-starved GHRY.5 cultures cxposed to GH for 2 h or less (Fig. 3B). Duplicate cultures exposed to GH for IO min, 30 min, 1 h or 2 h and then incubated with serum-free medium for a total incubation time of 7 h (the final 2 h with [3H]leucine) showed no increase in protein synthesis whereas cultures exposed to GH for the

B

-10 0

time Fig. 3. Stimulation

74 h serum deprivation.

cell cultures (A GHR9.5

and B) were pulsed with [‘H]leucine

TC’A precipitahle wcrc

protein\. incubated

cells by oGH

0.17

oGH

0.5

) were

and oPL. A: After 24 h serum deprivation,

incubated

in serum-free

cell cultures were cultured incubation

with serum-free

for the final 7 h of incubation or oPL.

growth medium with oGH

times. for a total incubation

1

2

GHR9.S

( *SD)

over

cell cultures (0

(0 and A ) or oPL (0).

medium containing oGH synthesis monitored

7

(h)

which was replaced

time of 7 h for all treatments.

and protein

The data are expressed as mean percent increase without

, 0

time

medium after the indicated

with strum-free (A

CHO

!

IO

(h)

of protein synthesis in GHR9.S

and 01 or non-transfected N: After

8

6

4

2

The cell cultures

by the incorporation

the controls where GHR9.5

* ,, < 0.01 compared with control (one-way analysis of variance).

cell cultures

of

43

these cells to a level detectable analysis.

by Northern

blot

5

679 -205

20+

Biochemical characterisation of GHR9.5 cells Affinity cross-linking analysis, using [ ‘2’I]oGH, of the expressed GH receptor on GHR9.5 cells shows a minor band at relative molecular weight (M,) 170,000 ( f 5000, number of evaluations (n) = 4) and a major band at M, 95,000 (k4000, n = 4); electrophoresis in the absence of pmercaptoethanol did not alter the size nor the relative distribution of these bands (Fig. 5A). The same analysis performed with sheep liver microsomes, in the presence of P-mercaptoethanol, revealed a similar pattern of bands; however, the minor and major bands were M, 155,000 ( f 4000, II = 2) and M, 80,000 ( k 3000, n = 2) respectively (Fig. 5B). Identical bands were found using either [“‘I]oGH (Fig. 5) or [““I]oPL (in GHR9.5 cells only, not shown), and addition of 1 pg of cold oGH resulted in no visible bands in either [ ‘251]oGH or [ ‘2”I]oPL cross-linking studies. No bands were visible when non-transfected CHO

d 11+

.116 -97

97-

.66

66-

.45 45-

.29 29_ 2_"G

+

+

-

-

+

+ t

t

008

-

+

-

+

+

_+

-

Fig. 5. Affinity cross-linking analysis of GHR9.5 cells and sheep liver microsomes. GHR9.5 cells (lanes l-6) or sheep liver microaomes (lanes 7 and 8) were incubated with [“‘I]oGH, in the presence or absence of I pg biological oGH. After washing, bound GH was cross-linked using DSS and solubilised in the presence (lanes I, 2, S-8) or absence (lanes 3 and 4) of @-mercaptoethanol and electrophoresed on 8% polyacrylamide gels. Dried gels were then autoradiographed for I day (lanes l-4) or 7 days (lanes S-8).

cells were cross-linked (not shown).

to these

labeled

ligands

Discussion

Fig. 4. Expression of oGH receptor and c-fos mRNA in GHR9.5 cells exposed to oGH. Total mRNA (10 pg) samples were electrophoresed on formaldehyde/agarose gels, transferred to Hybond Nf membranes and hybridised to GH receptor cDNA (A) or c-fos (B) probes. - FCS: serum-starved (quiescent) cells; + FCS: quiescent cells stimulated with medium containing 10% FCA, + FCS+ GH: quiescent cells stimulated with medium containing 10% FCS and 1 pg/ml oGH, + GH: quiescent cells stimulated with medium containing 1 fig/ml oGH.

The characterisation of GH receptors from a number of species has revealed surprising diversity in receptor structure (Mathews, 1991). It has been suggested that structurally discrete subpopulations of GH receptors may mediate the different physiological responses elicited by GH within specific tissues (Chipkin et al., 1989). In the present study CHO cells stably transfected with an ovine cDNA clone encoding a full-length, transmembrane GH receptor express on their cell surface a single class of high-affinity GH receptor. In the presence of ligand these receptors mediate at least one physiological response characteristic of GH action, the induction of protein synthesis.

44

Ligand-binding studies have revealed apparent heterogeneity with respect to the expression of somatotrophic receptors on ovine hepatic membranes. A single class of high-affinity receptor (K, = 0.27 nM) has been previously defined, the developmental expression of which is restricted to the postnatal animal (Freemark et al., 1986). Subsequently, both high-affinity (Kd = 0.17-0.31 nM) and low-affinity (Kd = 3-13 nM) somatotropic receptors have been detected as co-expressed on postnatal ovine hepatic membranes (Sauerwein et al., 1991). Significantly, the affinity coefficients calculated for the high-affinity class of receptor characterised in both studies agree well with that obtained by Scatchard analysis of oGH receptor expression by the stably transfected GHR9.5 cell line described here (Kd = 0.30 nM). This data, together with our previous observation that the oGH receptor cDNA used in the present study corresponds to an mRNA transcript that is predominantly expressed in postnatal liver (Adams et al., 1990) argues that all three high-affinity receptors described are equivalent, and encoded for by the same gene. No evidence was obtained from Scatchard analysis for the presence of a low-affinity class of GH receptors on GHR9.5 cells, and thus the structural relationship of this receptor to the high affinity receptor remains unclear. Low-affinity hepatic GH receptors have been described for a number of species including rats (Baxter et al., 19801, cattle (Brier et al., 1988) and pigs (Brier et al., 1989). Whether they represent post-translational modifications of high-affinity GH receptor translation products or are encoded by one of a number of alternatively spliced mRNA transcripts derived from a single GH receptor gene is not known. Their physiological significance is undefined. Cross-linking studies of GHR9.5 cells with I25I-labeled oGH reveals two bands of M, 170,000 and 95,000 (Fig. 5). Allowing for the recent observation that a single GH molecule has the capacity to bind one or two GH receptor molecules (Cunningham et al., 1991) the adjusted M, for the oGH receptors expressed by GHR9.5 cells becomes 148,000 and 73,000. Given that the fulllength GH receptor cDNA transfected into these cells codes for a protein with a calculated M, of 69,914, with seven potential N-linked glycosyla-

tion sites (Adams et al., 1990), then the M, 73,000 receptor on GHR9.5 cells presumably represents the full-length, transmembrane oGH receptor monomer. By extrapolation, the M, 148,000 species should represent a homodimeric form of the receptor cross-linked to a single oGH molecule. Parallel cross-linking analysis of endogenous oGH receptors reveals a corresponding pattern of receptor species, with adjusted molecular weights of 133,000 and 58,000 (Fig. 5). This would indicate an actual difference in the M, between endogenous and transfected oGH receptor monomers of 15,000. Affinity cross-linking has previously been used to characterise two species of somatotropic receptors expressed on postnatal sheep liver microsomes, one with an M, of 53,500 and a second receptor of M, 118,000 (Freemark et al., 1987). In contrast to our findings in the GHR9.5 cells, the high molecular weight oGH receptor described by Freemark and colleagues is sensitive to sulphydryl reducing agents, indicating assembly from disulphide-linked subunits (Freemark et al., 1987). This discrepancy between the studies remains unresolved. What is clear from studies on the biosynthesis of membrane-bound GH receptors performed in other species is that extensive post-translational modification of the receptor may occur, giving rise to extensive heterogeneity of membrane-bound GH receptors. Hocquette et al. (1990>, using cross-linking analysis, propose three or four membrane-bound forms of human liver GH receptor while six species of cross-linked receptors are seen in the rat (Moldrup et al., 1990). Differences in tissuespecific processing may explain, in part, the difference in size observed for oGH receptors expressed on transfected CHO cells when compared to endogenous hepatic receptors. The overexpression of rat GH receptors in CHO cells, following transfection with a cDNA clone encoding a full-length rat GH receptor, results in these cells acquiring a GH-responsive state, characterised by the stimulation of protein synthesis following GH exposure (Emtner et al., 1990). Similarly, oGH stimulates protein synthesis in GHR9.5 ceils overexpressing the oGH receptor (Fig. 3A), although the level of protein synthesis stimulated (27% increase relative to oGHunexposed controls) is less than the 60% induc-

45

tion observed in the previously cited study. With respect to the ability of GH to induce protein synthesis, CHO cells expressing rat or ovine GH receptors share a number of features. Transfected cells display GH responsiveness only after serum starvation, and require prolonged exposure to GH (9 h or more) before maximum stimulation of protein biosynthesis is observed (Fig. 3A; Emtner et al., 1990). Indeed, pulsing GHR9.5 cells for 10 min to 2 h with oGH, followed by an additional 5 h of culture does not augment protein biosynthesis over control levels (Fig. 3B). Prior exposure to GH is known to induce a period of refractoriness to subsequent exposure to GH with respect to the stimulation of amino acid uptake and protein synthesis (reviewed by Isaksson et al., 1985). As suggested by Emtner et al. (1990), FCS used for cell culture is a rich source of not only bGH but a variety of other growth factors whose potent metabolic effects include the stimulation of protein synthesis. Serum starvation, therefore, appears to be a necessary prerequisite before a window can be established for GH-induced protein synthesis to be observed. A contrasting observation is the ability of GH to rapidly stimulate protein synthesis in rat hemidiaphragms in vitro, where significant stimulation of protein synthesis is seen in tissue incubated with GH for as little as 10 min (Albertsson-Wikland et al., 1980a). This parallels the effect seen on protein synthesis in isolated rat hemidiaphragms taken from animals following GH administration in vivo (Albertsson-Wikland et al., 1980b) and most obviously indicates differences in the cellular assay systems being used. The mechanism(s) by which the GH receptor mediates the diverse physiological effects associated with GH is not known. Purified GH receptors isolated from a variety of cell types co-purify with a tyrosine kinase activity (Stred et al., 1990) although the receptor itself is not a tyrosine kinase. One of the earliest events associated with GH action in a number of cell types is the rapid, transient activation of the c-fos gene (Doglio et al., 1989; Gurland et al., 1990; Slootweg et al., 1990). In contrast to these results, we did not see any induction of c-fos mRNA expression in GHR9.5 cells by GH alone, although the addition of 10% FCS to serum-starved GHR9.5 cells char-

acteristically stimulated the appearance of c-fos mRNA transcripts. While GH appears to activate c-fos expression in preadipocytes and osteoblasts by independent pathways (Doglio et al., 1989; Slootweg et al., 1990) it is not known what role, if any, c-fos activation by GH may play in the stimulation of protein synthesis. The interaction of oPL with the oGH receptor merits particular interest. Amongst the placental lactogens isolated from a number of species, oPL appears unique in that it displays potent somatotropic activity, both in vivo (Ghan et al., 1976; Hurley et al., 1977) and in vitro (Freemark and Handwerger, 1982, 1983, 1984). This may be explained, in part, by the ability of oPL to bind GH receptors from a number of species (Carr and Friesen, 1976; Freemark et al., 1986). Indeed, oPL binds endogenous hepatic oGH receptors with an affinity equal to that of oGH (Freemark et al., 1986) despite displaying only 28% identity at the amino acid level (Colosi et al., 1989). Using E. c&-derived, recombinant oPL, we demonstrate unequivocally that oPL binds to oGH receptors expressed by the GHR9.5 cell line with an affinity slightly less than that observed for GH (K, oPL = 0.76 nM vs. K, oGH = 0.30 nM), with both hormones detecting a similar number of binding sites (- 200,000) per cell. Strikingly, oPL is able to stimulate a significant induction of protein synthesis in serum-starved GHR9.5 cells (38% increase relative to control cells), with a time-course of induction superimposable over that observed for oGH. Thus oPL is able to mediate a GH-like response through a specific interaction with a transfected GH receptor. What is intriguing about this observation is that specific, high-affinity hepatic oPL receptors have been identified in the sheep (Freemark et al., 1986). The ontogeny of expression of these receptors differs markedly from that of oGH receptors in that they appear as early as day 70 of gestation, and persist on a variety of tissues in postnatal animals (Ghan et al., 1978; Freemark et al., 1986). Interestingly, oGH binds oPL receptors with 60-fold less affinity compared to oPL, and displays little or no biological activity on fetal tissues expressing PL receptors (Hurley et al., 1980; Adams et al., 1983; Freemark and Handwerger, 1983,1986). Why oPL has evolved to retain its ability to mediate soma-

46

tropic effects through oGH receptors, in addition to oPL receptors, remains to be answered. Acknowledgements We wish to thank Dr. Andrew Nash for reviewing the manuscript, Ken Snibson for photography and Christine Kerr for typing the manuscript. Rodney J. Fiddes is a recipient of an Australian Wool Corporation postgraduate scholarship. This work was also supported by a Core Project grant from the Meat Research Corporation. References Adams, SO., Nissley. S.P.. Handwerger, S. and Rechler. M.M. (1983) Nature 302, 150-152. Adams, T.E., Baker, L.. Fiddes, R.J. and Brandon. M.R. (1990) Mol. Cell. Endocrinol. 73, 135-145. Albertsson-Wikland, K.. Eden, S., Ahren, K. and Isaksson. 0. (1980a) Endocrinology 106, 29X-305. Albertsson-Wikland. K.. Eden. S. and Isaksson, 0. (1980b) Endocrinology 106, 291-297. Barnard, R., Bundessen. P.G., Rylatt, D.B. and Waters, M.J. (IOXSI Endocrinology 115, 1805-1813. Bass, J.J., Oldham, J.M.. Hodgkinson, S.C., Fowke. P.J., Sauenvein, H., Molan. P., Brier, B.H. and Gluckman, P.D. (1991) J. Endocrinol. 128. 181-186. Baumann, O., Stolar, M.W., Amburn, K., Barsano, C.P. and De Vries, B.S. (1986) J. Clin. Endocrinol. Metab. 62, 134-141. Baumbach. W.A., Horner, D.L. and Logan, J.J. (lY8Y) Genes Dev. 3, 1199-1205. Baxter. R.C., Bryson, J.M. and Turtle. J.R. (1980) Endocrinology 107, 11761181. Baxter, R.C., Zaltsman. Z. and Turtle, J.R. (1984) Endocrinology 114, I893- 1903. Billestrup. N.. Moldrup, A., Serup, P., Mathews, L.S., Norsdedt, G. and Nielson, J.H. (1990) Proc. Natl. Acad. Sci. USA 87. 7210-7214. Boutin, J.M., Edery, M., Shirota. M., Jolicoeur, C., Lesuer, L., Ali, S., Djiane, L. and Kelly, P.A. (1989) Mol. Endocrinol. 3, 1455-1461. Brier, B.H., Gluckman, P.D. and Bass, J.J. (1988) J. Endocrinol. 123, 25-31. Brier. B.H.. Gluckman, P.D., Blair, H.T. and McCutcheon, S.N. (1989) J. Endocrinol. 123, 25-31. Carr. D. and Friesen, H.G. (1976) J. Clin. Endocrinol. Metab. 42, 4844493. Carter-%. C., Stubbert. J.R., Wang, X., Stred, S.E.. Argetsinger, L.S. and Shafer, J.A. (1989) J. Biol. Chem. 264, 18654- 1866 I,

Chart. J.S.D., Robertson. H.A. and Friesen. H.G. (1976) docrinology 98, 6576. Chart, J.S.D., Robertson, H.A. and Friesen, H.G. (1978) docrinology 102, 632-640. Chipkin, S.R., Szecowka. J., Tai, L.-H., Kostyo, J.L. Goodman, H.M. (1989) Endocrinology 12.5, 450-458. Chomczynski, P. and Sacchi, N. (1987) Anal. Biochem. 1.56~159.

EnEnand 162,

Colosi. P., Thordarson, G., Hellmiss, R.. Singh. K., Forsyth. I.A., Gluckman. P. and Wood, W.I. (1989) Mol. Endocrinol. 3. 146221469. Cunningham. B.C., Vltsch, M.. de Vos, A.M., Mulkerrin, M.G.. Clauser, K.R. and Wells. J.A. (1991) Science 254, 821-825. Davis. J.H. and Linzer, D.I.H. (1989) Mol. Endocrinol. 3, 674-680. Doglio, A., Dani, C.. Grimaldi, P. and Ailhaud. G. (1989) Proc. Natl. Acad Sci. USA 86, 1148-l 152. Emtner, M.. Mathews, L.S. and Norstedt. G. (1990) Mol. Endocrinol. 4, 2014-2020. Foster. C.M., Shafer, J.A., Rozsa, F.W., Wang, X.. Lewis. S.D.. Renken. D.A.. Natale, J.E., Schwartz, J. and CarterSu, C. (1988) Biochemistry 27, 326-334. Freemark, M. and Handwerger. S. (1982) Endocrinology 110, 2201-2203. Freemark, M. and Handwerger, S. (1983) Endocrinology 112, 402-404. Freemark, M. and Handwerger. S. (lY84) Am. J. Physiol. 246, E21-E24. Freemark, M. and Handwerger, S. (1986) Endocrinology 11X. 613-61X. Freemark. M., Comer, M. and Handwerger, S. (1986) Am. J. Physiol. 25 I, E32X-E333. Freemark. M.. Comer, M., Korner. G. and Handwerger, S. (1987) Endocrinology 120. 1865 1872. Friedman, J.S.. Cofer. C.L., Anderson. C.L., Kushner, J.A.. Gray, P.P.. Chapman. M.C., Stuart. M.C., Lazarus, L.. Shine, J. and Kushner. P.J. (1989) Bio/Technology 7, 3599362. Gluckman. P.D., Butler, J.H. and Elliot, T.B. (1983) Endocrinology 112, 1607-1612. Gluzman, Y. (1981) Cell 23. 175-182. Godowski, P.J., Leung, D.W., Meacham, L.R., Galgani, J.P., Hellmiss, R., Kenet, R., Rotwein. P.S., Parkes, J.S., Laron, Z. and Wood, W.I. (1989) Proc. Natl. Acad. Sci. USA 86, X083-8087. Gurland, K.. Ashcom, S., Cockram, B.H. and Schwartz. J. (1990) Endocrinology 127, 3187-3195. Herington, A.C., Ymer, S., Roupas, P. and Stevenson, J. (1986) Biochim. Biophys. Acta 881. 236-240. Hocquette, J.M., Postel-Vinay, M.-C., Djiane, J.. Tar, A. and Kelly, P.A. (1990) Endocrinology 127, 1665- 1672. Hurley, T.W., D’Ercode, A.J., Handwerger, S., Underwood, L.E.. Furlanetto, R.W. and Fellows, R.E. (1977) Endocrinology 101, 1635-1638. Isaksson, O.G.P., Eden, S. and Jansson. J.O. (1985) Annu. Rev. Physiol. 47, 4833499. Laemmli, V.K. (1970) Nature 2277, 680-685.

47 Leung, D.W., Spencer, S.A., Cachianes, G., Hammonds, R.G., Collins, C., Henzel, W.J., Barnard, R., Waters, M.J. and Wood, W.I. (1987) Nature 330, 537-543. Mathews, L.S. (1991) Trends Endocrinol. Metab. 2, 176-180. Moldrup. A., Billestrup, N. and Nielsen, J.H. (1990) J. Biol. Chem. 265, 8686-8690. Moldrup, A., Allevato, G.. Dyrberg, T., Nielsen, J.H. and Billestrup, N. (1991) J. Biol. Chem. 266, 17441-17445. Mountford, P.S. (1991) Isolation and Expression of DNA Sequences Encoding Ovine Follicle-Stimulating Hormone. Ph.D. Thesis, University of Melbourne. Melbourne. Mulligan, R.C. and Berg, P. (1981) Proc. Natl. Acad. Sci. USA 78, 2072-2076. Roupas, P., Chou, S.Y., Towns, R.J. and Kostyo, J.L. (1991) Proc. Natl. Acad. Sci. USA 88, 1691-1695. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual, 2nd edn., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Sauerwein, H., Breier, B.H., Bass, J.J. and Gluckman, P.D. (1991) Acta Endocrinol. 124, 307-313. Scatchard, G. (1949) Annu. NY Acad. Sci. 51, 660-672. Slootweg. M.C., Genesen, S.T., Otte, A.P., Duursma, S.A. and Kruijer, W. (1990) J. Mol. Endocrinol. 4, 265-274. Smith, W.C., Linzer, D.I.H. and Talamantes, F. (1988) Proc. Natl. Acad. Sci. USA 85, 9576-9579. Stredt, S.E., Stubbert, J.R., Argetsinger, L.S.. Shafer, J.A. and Carter-Su, C. (1990) Endocrinology 127. 2506-2516. Van Doren, K., Hanahan. D. and Gluzman, Y. (1984) J. Viral. 50, 606-614. Warwick, J.M. and Wallis, M. (1984) Biochem. Sot. Trans. 12, 247-248. Wigler, M., Sweet, R., Sim, G.K., Wold, B., Felliar, A., Lacy, E., Manniatis. T., Silverstein, S. and Axel, R. (1979) Cell 16. 777-785. Ymer, S.I. and Herington. A.C. (1985) Mol. Cell. Endocrinol. 41, 1533161.