0013-7227/92/1303-1483$03.00/0 Endocrinology Copyright 0 1992 hy The Endocrine

Vol. 130, No. 3 Printed in U.S.A.

Society

Regulation of Insulin-Like Growth Factor-Binding Protein Expression by Thyroid Hormone in Rat GHs Pituitary Tumor Cells* GIAN PAOLO CEDA, RON G. ROSENFELD,

PAUL AND

J. FIELDER?, SHARON ANDREW R. HOFFMAN

M. DONOVAN,

Medical Service, Department of Veterans Affairs Medical Center, Palo Alto, California 94304; and the Departments of Medicine and Pediatrics, Stanford University Medical School, Stanford, California 94305

ABSTRACT. Rat pituitary GH3 tumor cells express GH and insulin-like growth factor-I (IGF-I) mRNAs and possess specific IGF-I and IGF-II receptors which mediate the inhibitory effect of IGF on GH secretion. Ts increases the rate of cell replication and GH gene transcription, and causes an increase in IGF-I binding to cell membranes. Since the IGFs circulate in association with specific binding proteins (IGFBPs) that appear to modulate the access of IGFs ta their receptors, we have investigated the effect of T3 treatment on the expression of IGFBPs in GH, cells. Cells were grown in serum-free defined medium, and IGFBP secretion was determined by Western ligand blotting of conditioned medium after the addition of TI and/or various peptides for 48-72 h. The conditioned medium of GH, cells revealed a complex of bands migrating at 40-45 Mr, a pattern typical of rat (r) IGFBP-3. Ta treatment resulted in an increase in rIGFBP-3. IGF-I, IGF-II, and insulin did not alter rIGFBP3 levels. After concentrating (lo-fold) conditioned medium samples, two additional bands at 24K and 28K mol wt were also seen. These bands corresponded in size to rIGFBP-4 (24K) and its glycosylated form (28K). The mRNAs for both rIGFBP-3

and rIGFBP-4 were present in GHS cells; TS treatment increased steady state levels of rIGFBP-3 mRNA, but did not affect BP-4 mRNA levels. To learn whether the increased expression of IGFBPs could be responsible for the increased IGF-I biding seen aftar T3 treatment, [?]IGF-I was cross-linked to GHI membranes, and the proteins were separated on a 5-E% gradient sodium dodecyl sulfate-polyacrylamide gel under reducing conditions. Ts treatment induced a large increase in the intensity of bands migrating at 135K and 280K, representing the cr-subunit of the IGF-I receptor and an incompletely reduced cu-a-dimer, respectively. No membrane-associated IGFBPs were detected. In conclusion, GHs cells express two IGFBPa, rIGFBP-3 and rIGFBP-4, which are differentially regulated by T,. The increased binding of IGF-I to GHB cell membranes after Ta treatment indicates that thyroid hormone induces an up-regulation of IGF-I receptors and that the increased IGF-I binding to GHB membranes is not due to increased expression of membraneassociated IGFBPs. (Endocrinology 130: 1483-1489,1992)

T

HYROID hormone is required for normal GH gene expression in vitro (1) and in uiuo (2). GH cells are clonal strains of rat pituitary tumor cells that synthesize and secrete GH (3) and contain multiple TB receptors (4). GHB cells have been extensively employed to study the hormonal regulation of GH synthesis and release and to investigate the role of insulin-like growth factors (IGFs) in the pituitary feedback regulation of GH secretion. Using this cell line, it has been shown that thyroid hormone increases GH gene transcription, protein synthesis, and hormone secretion (5, 6). In addition TB treatment increases the rate of replication of GH cells, acting early in the Gl phase of the cell cycle (7-9).

Moreover, the conditioned medium of Ta-treated GH cells stimulates the growth of somatotrophs (10-12). Together, these data indicate that TB may directly enhance the growth of somatotrophs in addition to inducing the production of one or more autocrine growth factors. We have previously demonstrated the presence of abundant specific IGF-I and IGF-II receptors in three GH cell lines (13). IGF-I mRNA is present in GHB cells, and these cells have been shown to secrete immunoreactive IGF-I (14). T3 increases IGF-I gene expression in these cells (14, 15). In addition, T3 increases the specific binding of IGF-I to GHB cells without altering the binding of either IGF-II or insulin (16). While IGFs exert their biological action by binding to specific cell surface receptors on target cells, the IGFs also bind to members of a family of circulating proteins that have high affinity and specificity for the IGFs, the IGF-binding proteins (IGFBPs) (17-19). The IGFBPs can modulate the proliferative and mitogenic effects of

Received July 22,1991. Address all correspondence and requests for reprints to: Andrew R. Hoffman, M.D., Medical Service (ill), Veterans Administration Medical Center, 3801 Miranda Avenue, Palo Alto, California 94304. * This work was supported by the Research Service of the Department of Veterans Affairs and NIH Grant DK-36054. t Supported by National Research Scientist Award DK-08428. 1483

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.

1484

THYROID

HORMONE

IGFs in many biological systems (17). The production of IGFBPs by pituitary tissue was first demonstrated by Binoux et al. (20) in explants of rat anterior pituitary and neurointermediate lobes. A small IGFBP has been identified in human pituitary tissue, and more recently, Rosenfeld et al. (21) confirmed the presence of heterogeneous low mol wt IGFBPs in the conditioned medium of cultured rat anterior pituitary and neurointermediate cells using competitive binding, affinity cross-linking, and Western ligand blot techniques. Since GH3 cells have been shown to secrete IGFBPs (21) into the culture medium and because these IGFBPs may be important modulators of IGF action in the pituitary, the aim of this study was to characterize the IGFBPs secreted by GH3 cells, to study the regulation of IGFBP expression by TB, and to determine whether the Ta-induced increase in IGF-I specific binding to GH, cell membranes is due to an increased expression of IGFBPs or IGF-I receptors. Materials

and Methods

Materials

Synthetic human IGF-II was kindly provided by the late C. H. Li (San Francisco, CA). Recombinant DNA-derived [Thr59] IGF-I was purchased from Amgen Biologicals (Thousand Oaks, CA). Peptides were iodinated by a modification of the chloramine-T method to a specific activity of 150-300 rCi/pg for both peptides. Bovine insulin was supplied by Eli Lilly Co. (Indianapolis, IN). TS and dexamethasone were obtained from Sigma (St. Louis, MO). Cell culture

GHs cells were obtained from the American Tissue Type Culture Collection (Rockville, MD). Cells were maintained in a 1:l mixture of Dulbecco’s Modified Eagle’s Medium-Ham’s F-12 (DMEM-F12; Sigma) enriched with fetal calf serum (FCS2.5%), horse serum (15%), penicillin (100 U/ml), and streptomycin (100 rg/ml) in 75-cm flasks. FCS was stripped of thyroid hormones and glucocorticoids using a previously described procedure (22). Cultures were passaged by trypsinEDTA treatment every 7-10 days (1:4 split). For hormonal regulation experiments, cells were plated in 24-multiwell plates in DMEM-F12 with 10% hormone-stripped FCS and allowed to attach to the dish for 48 h. The medium was subsequently

discarded and replaced with fresh serum-free chemically defined medium (SFM) containing transferrin (100 kg/ml), sodium selenite (30 nM), and putrescine (100 PM) for an additional 24 h. Subsequently, the medium was discarded and replaced with the serum-free chemically defined medium containing the various experimental hormones. Conditioned medium samples were collected 48 or 72 h after the hormones were added to the cells. The conditioned medium was centrifuged at 2000 x g for 10 min at 4 C to remove cellular debris. At the time of medium collection, cell counts of the corresponding GHs monolayer were determined with a Coulter counter (Hialeah, FL). Conditioned medium was also collected from BRL3A2 cells (known

REGULATION

OF IGFBP

Endo. 199’ Vu1 130. No 3

to secrete predominantly rIGFBP-2), from H35 cells (which secrete both rIGFBP-1 and 24K IGFBP), and from B-104 cells (which predominantly secrete rIGFBP-4), grown as previously described (21, 23). For membrane preparation, GH, cells were grown to confluence in a 1:l mixture of DMEM-F12 (Sigma) enriched with FCS (2.5%), horse serum (15%), penicillin (100 U/ml), and streptomycin (100 fig/ml) in 150-cm flasks. The medium was then aspirated, and SFM, with or without 50 nM Ta, was added for 48 h. The cells were subsequently washed three times with PBS, detached with 1 mM EDTA, and then disrupted by sonication for l-2 min at 4 C. The lysate was centrifuged at 4,000 x g for 30 min at 4 C. The pellet was discarded, and the supernatant was collected and centrifuged at 40,000 x g for 1 h at 4 C. The resulting pellet was resuspended in 5 mM Tris buffer with 1 mM phenylmethylsulfonylfluoride and frozen at -20 C for subsequent experimental use. Affinity

cross-linking

Cell membranes (1 mg/ml) were incubated with [‘251]IGF-I (2 x lo5 cpm) and increasing concentrations (1, 10, 100, and 500 rig/ml) of unlabeled IGF-I and IGF-II. The samples were incubated at 25 C in a final volume of 1 ml for 3 h. The samples were then washed with 1 ml binding buffer and centrifuged at 4500 x g for 30 min at 4 C. The pellet was resuspended in 1 ml 5 mM Tris buffer without BSA, and the radiolabeled peptide was cross-linked to its receptor with 10 ~11 rnM disuccinimidyl suberate in 100% dimethylsulfoxide (pH 7.4) at 4 C. The crosslinking reaction was quenched after 15 min by the addition of 200 ~150 mM Tris-HCl and 5 mM EDTA, followed by a second centrifugation. The pellet was then solubilized in sodium dodecyl sulfate (SDS) sample buffer containing 5% P-mercaptoethanol, boiled for 5 min, and subsequently electrophoresed on a 5-15% acrylamide separating gel using constant current. The gels were vacuum dried at 80 C for 2 h and exposed to Kodak XAR-2 film (Eastman Kodak, Rochester, NY). SDS-polyacrylamide ligand blotting

gel electrophoresis

(SDS-PAGE)/

Western

SDS-PAGE was performed according to the method of Laemmli (24). GHB-conditioned medium was applied to a 4% stacking gel and electrophoresed through a 10% or 12.5% polyacrylamide gel. Prestained mol wt standards (Bethesda Research Laboratories, Gaithersburg, MD) were run in parallel lines. Some IGFBP samples were enzymatically deglycosylated using endoglycosidase-F (Calbiochem, La Jolla, CA) as follows. Samples were adjusted to pH 5.5, and 240 mU endoglycosidaseF were added. After incubation at 37 C for 3 h, the reaction was terminated by the addition of an equal volume of electrophoresis buffer. Gels were run overnight at 50 V. After electrophoresis, polyacrylamide gels were washed in Towbin buffer (0.025 M Tris-base, 0.192 M glycine, and 20% methanol) for 15 min to remove SDS before electrotransfer. Proteins were electroblotted onto nitrocellulose (0.45 @cm; Schleisher and Schuell, Inc., Keene, NH) using a Gelman Biotrans semidry electrophoresis transfer unit (Gelman Sciences, Ann Arbor, MI) with Towbin buffer. Western ligand blotting was carried out by the method of Hossenlopp et al. (25). Nitrocellulose membranes were washed with TBS (0.15 M

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.

THYROID

HORMONE

sodium chloride and 0.01 M Tris-HCl) containing 3% Nonidet P-40 (Sigma). Membranes were blocked with TBS containing 1% BSA (Sigma) for 2 h, followed by TBS containing 0.1% Tween for 10 min. The membranes were then incubated overnight at 4 C with approximately 1.5-2 x lo6 cpm [1251]IGF-I or [12SI]IGF-II in 20 ml TBS containing 1% BSA and 0.1% Tween. Membranes were washed with TBS, air dried, and visualized by exposure to Kodak X-Omat AR film for 3-7 days at -70 C.

REGULATION

OF IGFBP

1485

ABCDEFGHI I I I

-45K

-

I

I

I

I

I

I

/I

RNA isolation and analysis Total cellular RNA was isolated from cells by the guanidine thiocyanate-LiCl precipitation technique (26). The precipitated RNA was pelleted by centrifugation, washed with cold 3 M LiCl, and dissolved in Tris-SDS buffer. The RNA was deproteinized by phenol-chloroform and chloroform extraction, and then precipitated with ethanol. Twenty micrograms of total RNA were loaded onto each lane. The RNA was size fractionated on 1.2% agarose-formaldehyde gels (27), transfered to nitrocellulose, and then hybridized with a 32P-labeled l.l-kilobase (kb) EcoRI-P&I 5’-fragment of human IGFBP-3 cDNA that includes the entire coding sequence (kindly provided by W. Wood, Genentech, South San Francisco, CA) (28), a 1.2-kb fragment of human IGFBP-4 cDNA that includes the entire coding sequence (a kind gift of S. Mohan, Loma Linda, CA) (29), or a chicken @actin probe to assessthe relative amounts of RNA loaded onto each lane. A RNA ladder (Bethesda Research Laboratories) was run for RNA sizing. The probes were labeled with 32P by random oligo priming (Pharmacia, Uppsala, Sweden) and purified on spin columns (Bio-Rad, Richmond, CA). They each had specific activities of approximately 2 x 10’ dpm/pg. Hybridized filters were washed for two lo-min periods at 55 C in 0.2 X SSC (1 X = 150 mM sodium chloride and 15 mM sodium citrate, pH 7.0)-0.05% SDS, air dried, and then exposed to XAR film with intensifying screens at -70 C. The blot was first hybridized to the IGFBP-4 probe, and then stripped by soaking the nitrocellulose twice in hot boiled diethylpyrocarbonate-treated water for 15 min each time; the filters were rinsed with hot diethylpyrocarbonate-treated water between soaks. The stripped blots were then soaked in prehybridization buffer and reprobed with the IGFBP-3 cDNA. Finally, the stripping process was repeated, and the actin probe was applied. Den&m&y Autoradiograms were scanned using laser densitometry (Ultrascan XL, LKB, Bromma, Sweden). The absorbance units (AU) per mm were measured for each band and used to determine the total binding activity per lane and the relative proportion of the binding activity associated with each band.

Results The Western ligand blot of IGFBPs in the concentrated conditioned medium of GHB cells is shown in Fig. 1. In the conditioned medium of cells incubated without any hormone added, a complex of bands migrating at 4045K mol wt was visualized (Fig. 1, lane E). These bands correspond to the predominant IGF-binding component

=29~-

1

=24K-:

FIG. 1. Western ligand blot analysis of IGFBPs in the conditioned medium of BRL3A cells (lane A), B-104 ceils (lanes B-C), rat serum (lane D), and GH, cells (lanes E-I). After plating, GH, cells were incubated for 48 h in SFM; the SFM was exchangedwith fresh SFM without TB (lane E) or SFM containing increasingconcentrations of TS [O.OlnM (lane F), 0.1 nM (lane G), 1 nM (lane H), and 10 nM (lane I)]. Conditioned medium was collected 48 h after the addition of TB and concentrated 10 times by Centriprep centrifugation. IGFBPs were separatedby SDS-PAGEon a 10% gel, transferred to nitrocellulose, ligand blotted with [‘251]IGF-II,and visualizedvia autoradiography.

present in adult rat serum (lane D), identified as rIGFBP-3. When conditioned medium from GH3 cells was concentrated 10 times by Centriprep centrifugation, two additional bands were identified: a major band migrating at 24K mol wt and a minor one at 28K, which correspond in size to rIGFBP-4 and its glycosylated form. Lanes B and C of Fig. 1 show intact rIGFBP-4 and its glycosylated form obtained from a partially purified preparation of conditioned medium of B-104 cells, which predominantly express both forms of rIGFBP-4, migrating at the same mol wt as the bands in GHa-conditioned medium. The identity of the 24K and 28K bands in the B104 cells was proven by partial sequence analysis (23). Therefore, the major IGFBP in the conditioned medium of GH3 cells is rIGFBP-3, with lesser amounts of BPS that migrate similarly to rIGFBP-4 also present. Addition of increasing concentrations of TS (from O.Ol10 nM; Fig. 1, lanes F-I) resulted in an apparent

dependent increase in rIGFBP-3 imal effect at lo-100

dose-

expression, with a max-

nM. Even at the highest doses of

TS employed, there was no significant change in the intensity of the bands migrating at 24K and 28K mol wt, corresponding to rIGFBP-4. Treatment of conditioned

medium

with endoglycosi-

dase-F reduced the size of the 40-45K IGF-binding components to two bands migrating at apparent mol wt of 38K and 34K, indicating that the large bands corresponding to rIGFBP-3 contain N-linked oligosaccharides (Fig.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.

THYROID

1486

HORMONE

REGULATION

OF IGFBP

Endo. 1992 Voll30. No 3

2A, conditioned media were not concentrated for this gel). To learn if the 28K IGFBP band is glycosylated, lo-fold concentrated conditioned medium was exposed to endoglycosidase-F. As shown in Fig. 2B, the 28K bands in both GH3 and B104 cells were dramatically decreased after enzyme treatment, suggesting that the 28K band represents glycosylated IGFBP-4. GHS cells were then treated with IGF-I (10 nM), IGFII (10 nM), or insulin (5 PM) with or without T3 (20 nM). As shown in Fig. 3, addition of IGF or insulin for 48 h did not alter the secretion of IGFBP-3, while TB, alone or in combination with the various peptides, caused a 3to Q-fold stimulation of IGFBP-3 expression. As shown in Fig. 4, there was a marked increase in the rIGFBP-3 mRNA concentration after T3 treatment. When RNA from GH3 or H35 cells was hybridized with a human IGFBP-4 cDNA probe, a single transcript with an apparent mol wt of 1.8 kb was identified; two trancontrol + T3 T3 + IGF-I RS

i

i

i

i

i

i

i

i 3. Densitometric analysis of Western liiand blot demonstrating the effects of IGF-I, IGF-II, and insulin, without and with the addition of Ta, on the expression of IGFBP-3 in GHa cells. Cells were plated in DMEM-F-12 with 10% hormone-stripped FCS for 48 h; medium was discarded and changed to SFM containing IGF-I (10 nM), IGF-II (10 nM), and insulin (5 PM) without and with Ts (20 nM). Conditioned medium was collected 48 h after the addition of the growth factors. IGFBPs were separated by SDS-PAGE on a 10% gel, transferred to nitrocellulose, ligand blotted with [‘*“I]IGF-II, and visualized via autoradiography. Relative quantities of rIGFBP-3 were determined after scanning the bands on the autoradiograph corresponding to 40-4513 mol wt with a laser scanning densitometer. Results are expressed as a percentage of the control values + SEM for three separate experiments, each performed in duplicate.

FIG.

=29K-

=24K-

A 8104 Mrx

10e3

i

GH3 +

I

+

i

RS

I

+

;

I

4629 -

18-

6 2. Autoradiograph of GH3 IGFBPs incubated at 37 C for 3.5 h in the absence (-) or presence (+) of endoglycosidase-F. RS represents the IGFBPs present in rat serum. Immediately after deglycosylation, IGFBPs were separated by SDS-PAGE on 10% gel, transferred to nitrocellulose, ligand blotted with [‘z61]IGF-II, and visualized via autoradiography. A, GHs cells were treated with SFM (control), TB (20 mu), or IGF-I (10 nM) plus TS (20 nM). B, Medium from GHI cells was concentrated IO-fold in order to visualize the low abundance 28K band.

FIG.

scripts of 2.4 and 1.8 kb mRNA were seen in adult rat liver. Treatment of GH3 cells with 50 nM T3 did not result in a significant change in the expression of IGFBP4 mRNA. To learn whether increased expression of IGFBPs (in particular IGFBP-3) was responsible for the increased specific binding of IGF-I to intact GHB cells after T3 treatment (16), [‘251]IGF-I was cross-linked to GHs membranes and separated on a 5-15% gradient SDS-PAGE under reducing conditions. As shown in Fig. 5, the predominant binding of [‘251]IGF-I was to two proteins, one with an apparent mol wt of 135K, representing the LYsubunit of the type 1 receptor, and one with a ma1 wt of 280K, presumably representing an incompletely reduced a-cu-dimer. Excess unlabeled IGF-I (200 ng) completely inhibited the binding of [12”I]IGF-I to both the a-subunit and the cY-a-dimer. Treatment with 50 nM T3 for 72 h induced a 200% increase in IGF-I binding to both proteins. No other bands could be visualized at lower mol wt corresponding to the identified IGFBPs. The binding

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.

THYROID C I

GH3 Cells C T3 I I

Ta 8104 H35 I I I

HORMONE

ICFBP-4

;

Kb

-1.8

Kb

-2.6

Kb

-1.8

Kb

FIG. 4. Northern blot demonstrating the effects of T3 on the expression of IGFBP-3 and IGFBP-4 in GH, cells. GH, cells were grown in SFM with or without (control; C) the addition of TB (50 mu) for 72 h. Total RNA was isolated, size-fractionated, transferred to nitrocellulose, and hybridized with a 8*P-labeled human IGFBP-3 cDNA probe, a chicken @-actin cDNA probe, or a human IGFBP-4 cDNA probe. Total RNA was also isolated from B-104 cells, the rat liver cell line H35, and adult rat liver (RL).

Control Bo +ICF-I I

II

T3 Treated Be II-i-i I

OF IGFBP

1487

Discussion

RL I

-2.6

Betr.Actin

REGULATION

+ICF-I

--280K

--130K

FIG. 5. Autoradiograph of [1*61]IGF-I cross-linked to GHI membranes (B.) and dieplaced with excess unlabeled IGF-I (200 ng) in control and Ta (50 m&treated cells. GHs cell membranes were incubated with [‘“IIIGF-I with or without unlabeled IGF-I. After incubation, receptor complexes were cross-linked with disuccinimidyl suberate, separated on 5-15% gradient SDS-PAGE under reducing conditions, and viaualized via autoradiography. Bo, Bound.

of [‘251]IGF-II to the type 2 IGF receptor was not affected by T3 treatment (data not shown).

The role of IGFs in somatotroph physiology has been extensively studied in uiuo (30, 31) and in vitro (31-39). We previously characterized specific receptors for insulin, IGF-I, and IGF-II on primary monolayer cultures of rat and human anterior pituitary cells, human somatotroph tumors, and GH cell lines. In GH3 cells, the f(d of the high affinity IGF-I receptor is -1 x lo-“, and there are approximately lo* binding sites/cell (13). Although GHB cells primarily secrete GH in a constitutive manner, it has been shown that IGF-I can inhibit GH synthesis and secretion from this cell line (39, 40). In addition, IGF-I increased DNA and protein synthesis in GH, cells. Thyroid hormone exerts a stimulatory effect on several aspects of the IGF system of the GH3 cell. T3 increased IGF-I binding to cell membranes in a dose- and timedependent manner without substantially altering IGF-II or insulin binding (16). Melmed et al. (41) also reported that thyroid hormone enhanced insulin binding to GH cells. Locally produced IGFBPs may also regulate GH secretion. Binoux et al. (20) demonstrated that IGFBPs were synthesized by rat pituitary glands. We previously characterized IGFBPs in conditioned medium from primary cultures of both rat anterior pituitary and neurointermediate lobe (21). When either [‘251]IGF-I or -11 was cross-linked to conditioned medium from anterior pituitary cultures, multiple small IGFBPs were identified. Treatment of anterior pituitary-conditioned media with endoglycosidase-F reduced the apparent mol wt of the predominant large mol wt BP, suggesting that the major BP in anterior pituitary was glycosylated IGFBP-3. In neurointermediate lobe-conditioned medium, the predominant IGFBP had an apparent mol wt of 2627K by ligand blot, was not glycosylated, and could be immunoprecipitated by antibody to IGFBP-2. In this study we have shown that the primary IGFBPs in GH3 pituitary cells are rIGFBP-3 and, to a much lesser extent, rIGFBP-4. The mRNAs for both IGFBP-3 and IGFBP-4 were present in GH3 cells. A single 2.6-kb IGFBP-3 transcript was seen in each cell type. As we have previously reported (23), in B104 ceils and rat liver, two IGFBP-4 mRNA transcripts were seen: a major band at 1.8 kb and a minor band at 2.6 kb. GH3 cells and rat hepatocyte H35 cells only express the 1.8-kb transcript, however. Recently, Cheung et al. (42), using a rat IGFBP4 probe, reported the presence of a single 2.6-kb IGFBP4 mRNA transcript in B104 cells. In GH3 cells, the secretion of rIGFBP-3 is increased by TB, but IGF-I, IGF-II, and insulin did not change the concentrations of rIGFBP-3 in conditioned medium. In addition, the rIGFBP-3 mRNA content was markedly stimulated by T3 treatment. On the other hand, the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.

THYROID

1488

HORMONE

concentration of rIGFBP-4 mRNA and the levels of the 24K and 28K BPS in the conditioned medium were not substantially influenced by T3. The different secretory patterns of the two IGFBPs present in the conditioned medium of GH3 cells after Ts treatment indicate different regulatory control of these two IGFBPs and suggest that these IGFBPs may have different effects on the modulation of GH secretion at the pituitary level. Since rIGFBP-3 levels are increased by Ta, it was important to learn whether the enhanced [1251]IGF-I binding that we previously reported in Ts-treated GH3 cells was caused by an increase in membrane-associated rIGFBP-3 or IGF receptors. By cross-linking [‘251]IGF-I to GH3 membranes, we demonstrated that the increase in membrane binding of [‘251]IGF-I is caused by an increase in the type I IGF receptor. Thus, T3 appears to have two discrete effects on IGF binding to GH, cells: an increase in the type I IGF receptor and an increase in rIGFBP-3. While no membrane-bound IGFBPs were detected by the method employed, it is possible that some IGFBPs can become associated with the cell surface. For example, IGFBP-3 can bind to the cell surface of cultured bovine fibroblasts, and when bound to the cells, the BP can enhance IGF-I binding (43) In this study BPS were measured in cells that were adherent to the plastic dishes, and cross-linking studies of the IGF-I receptor were performed in membranes derived from adherent cells. Since cell surface-associated BPS can potentially compete with IGF receptors for binding of the IGFs and thereby modulate IGF hormone action, it will be important to devise methods that can accurately identify and quantitate BPS on the surface of intact cells. The role of T3 in GH homeostasis is complex. In the absence of thyroid hormone, GH synthesis is severely diminished, and young hypothyroid animals have decreased linear growth. The content of immunoreactive GH-releasing hormone in the hypothalamus is decreased by 50% in hypothyroid rats, and there is decreased pituitary responsiveness to GH-releasing hormone administration (44, 45). T3 therefore increases the ability of the hypothalamic-somatotroph axis to increase GH synthesis and release (21), and IGF-I inhibits the ability of T8 to induce GH gene expression (40). Finally, T3 induces the synthesis of the major IGFBP in the somatotroph, rIGFBP-3. The physiological function of IGFBPs in the pituitary is not known, but it is possible that the BPS could modulate the negative feedback effects of IGF-I, preventing the locally synthesized IGF-I from inhibiting the synthesis and release of GH from the somatotroph. References 1. Shapiro LE, Samuels HM, Yaffe BM 1978 Thyroid and glucocorticoid hormones synergistically control growth hormone mRNA in cultured GHl cells. Proc Nat1 Acad Sci USA 75x45-49

REGULATION

OF IGFBP

Endo. 199’2 Vol130. No 3

2. Samuels MH, Wierman ME, Wang C, Ridgway EC 1989 The effect of altered thyroid status on pituitary hormone messenger ribonucleic acid concentrations in the rat. Endocrinology 124:2277-2282 3. Tashjian Jr AH 1979 Clonal strains of hormone-producing pituitary cells. Methods Enzymol58:527-536 4. Lazar MA. Chin WW 1988 Resulation of two c-e&A messeneer ribonucleid acids in rat GH3 celis by thyroid hormone. Mol Endocrinol 2:479-484 5. Crews MD, Spindler SR 1986 Thyroid hormone regulation of the transfected rat growth hormone promoter. J Biol Chem 261:50185022 6. Larsen PR, Harney JW, Moore DD 1986 Sequences required for cell-type specific thyroid hormone regulation of rat growth hormone promoter activity. J Biol Chem 261:14373-14376 7. DeFesi CR, Fels EC, Surks MI 1984 Triiodothyronine stimulates growth of cultured GC cells by action early in the Gl period. Endocrinology 114:293-295 8. DeFesi CR, Fels EC, Surks MI 1985 L-Triiodothyronine (T3) stimulates growth of cultured GC cells by action early in the Gl period: evidence for mediation by the nuclear T3 receptor. Endocrinology 116:2062-2069 9. Kitagawa S, Obata T, Willingham MC, Cheng SY 1987 Thyroid hormone action: induction of morphological changes and stimulation of cell growth in rat pituitary tumor GHI cells. Endocrinology 120:2591-2596 10. Danielpour D, Ikeda T, Kunkel MW, Sirbasku DA 1984 Identifcation of an autostimulatory growth factor in the extracts and conditioned medium of GH3/Cl4 rat pituitary cells in culture. Endocrinology 115:1221-1223 11. Miller MJ, Fels EC, Shapiro LE, Surks MI 1987 L-Triiodothyronine stimulates growth by means of an autocrine factor in a cultured growth hormone-producing cell line. J Clin Invest 79:1773-1781 12. Hinkle PM, Kinsella PA 1986 Thyroid hormone induction of an autocrine growth factor secreted by pituitary tumor cells. Science 2341549-1552 13. Rosenfeld RG, Ceda GP, Cutler CW, Dollar LA, Hoffman AR 1985 Insulin and insulin-like growth factor (somatomedin) receptors on cloned rat pituitary tumor cell lines. Endocrinology 117:2008-2016 14. Fagin JA, Pixley S, Slanina S, Ong J, Melmed S 1987 Insulin-like growth factor I gene expression in GH3 rat pituitary cells: messenger ribonucleic acid content, immunocytochemistry, and secretion. Endocrinology 120~2037-2043 15. Fagin JA, Fernandez-Mejia C, Melmed S 1989 Pituitary insulinlike growth factor I gene expression: regulation by triiodothyronine and growth hormone. Endocrinology 125:2385-2391 16. Geary ES, Lim M, Ceda GP, Ro S, Rosenfeld RG, Hoffman AR 1989 Triiodothvronine regulates insulin-like growth factor I binding to culturedrat pituitary cells. J Neuroend‘bcrinol 1:179-l&1 17. Rosenfeld RG, Lamson G, Pham H, Oh Y, Conover C, De Leon D, Donovan S. Ocrant I. Giudice L 1990 Insulin-like growth factor binding proteins. Recent Prog Horm Res 4699-163 18. Ballard J, Baxter R, Binoux M, Clemmons D, Hall K, Hintz R, Rechler M, Rutanen M, Schwander E 1989 On the nomenclature of the IGF binding proteins. Acta Endocrinol (Copenh) 121:751752 19. Lamson C, Giudice L, Rosenfeld RG 1991 Insulin-like growth factor binding proteins: structural and molecular relationships. Growth Factors 5:19-28 20. Binoux M, Hossenlopp P, Lassarre C, Hardouin N 1981 Production of insulin-like growth factors and their carrier by rat pituitary aland and brain exnlants in culture. FEBS Lett 124~178-184 21. Rosenfeld RG, Pham H, Oh Y, Ocrant I 1989 Characterization of insulin-like growth factor binding proteins in cultured rat pituitary cells. Endocrinology 1242867-2874 22. Samuels HH, Stanley F, Casanova J 1979 Depletion of L-3,5,3’triiodothyronine and- L-thyroxine in euthyroid-calf serum for use in cell culture studies of the action of thyroid hormones. Endocrinology 105:80-85 23. Ceda GP, Fielder PJ, Henzel WJ, Louie A, Donovan SM, Hoffman AR. Rosenfeld RG 1991 Differential effects of IGF-I and IGF-II on the expression of IGF binding proteins (IGFBPs) in a rat

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.

THYROID

24. 25.

26. 27.

28.

29.

30. 31.

32. 33. 34.

36.

HORMONE

neuroblastoma cell line: isolation and characterization of two forms of IGFBP-4. Endocrinology 1282815-2824 Laemmli V 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 Hossenlopp P, Seurin D, Sequoia-Quinson B, Hardouin S, Binoux M 1986 Analysis of senrm insulin-like growth factor binding proteins using Western blotting: use of the method for titration of binding proteins and competitive binding studies. Anal Biochem 164:138-143 Cathala G, Savouret JF, Medez B, West BL, Karin M, Martial JA, Baxter JD 1983 A method for isolation of intact, translationally active ribonucleic acid. DNA 2:329-335 Lehrach H, Diamond D, Wozney JM, Boedtker H 1977 RNA molecular weight determination by gel electrophoresis under denaturing conditions: a critical reexamination. Biochemistry 164743-4751 Wood WI, Cachianes G, Henzel WJ, Winslow GA, Spencer SA, Helmiss R, Martin JL, Baxter RC 1988 Cloning and expression of the growth hormone-dependent insulin-like growth factor binding protein. Mol Endocrinol2:1176-1186 La Tour D, Mohan S, Linkhart TA, Baylink DJ, Strong DD 1990 Inhibitory insulin-like growth factor-binding protein: cloning, complete sequence, and physiological regulation. Mol Endocrinol 4:18061814 Tannenbaum GS, Guyda HJ, Posner BI 1983: Insulin-like growth factors: a role in growth hormone negative feedback and body weight regulation via brain. Science 220:77-79 Abe H, Molitch ME, Van Wyk JJ, Underwood LE 1983 Human growth hormone and somatomedin C suppress the spontaneous release of growth hormone in unanesthetized rats. Endocrinology 113:1319-1324 Goodyer CG, DeStephano L, Guyda HJ, Posner PI 1984 Effects of insulin-like growth factors on adult male rat pituitary function in tissue culture. Endocrinology 115:1568-1576 Sheppard MS, Bala RM 1986 Insulin-like growth factor inhibition of growth hormone secretion. Can J Physiol Pharmacol 64:525530 Yamashita S, Melmed S 1986 Insulin-like growth factor I action on rat anterior pituitary cells; suppression of growth hormone secretion and messenger ribonucleic acid levels. Endocrinology 118176-182 Ceda GP, Narog B, Rosenfeld RG, Hoffman AR 1986 The role of

REGULATION

36.

37.

38.

39.

40. 41.

42.

43. 44.

45.

OF IGFBP

1489

insulin-like growth factors and insulin in the regulation of growth hormone secretion. In: Molinatti GM, Martini L (eds) Endocrinology ‘85. Elsevier, Amsterdam, pp 147-152 Goodyer CG, Marcovitz S, Hardy J, Lefebre L, Guyda H, Posner BI 1986 Effect of insulin-like growth factors on human fetal, adult normal and tumor pituitary function in tissue culture. Acta Endocrinol (Copenh) 112:49-57 Yamashita S, Weiss M, Melmed S 1986 Insulin-like growth factor I regulates growth hormone secretion and messenger ribonucleic acid levels in human pituitary tumor cells. J Clin Endocrinol Metab 63:730-735 Ceda GP, Davis RG, Rosenfeld RG, Hoffman AR 1987 The growth hormone (GH) releasing hormone (GHRH)-GH-somatomedin axis: evidence for rapid inhibition of GHRH-elicited GH release by insulin-like growth factors I and II. Endocrinology 120:16581662 Keiichi M, Yamashita S, Niwa M, Kurihara M, Harakawa S, Izumi M. Naeataki S. Melmed S 1990 Thvroid hormone reaulatas rat pit&a& insulin-like growth factor-1 receptors. En&crinology 126550-554 Yamasaki H, Prager D, Gebremedhin S, Moise L, Melmed S 1991 Binding and action of insulin-like growth factor I in pituitary tumor cells. Endocrinology 128857862 Melmed S. Yamashita S 1986 Insulin-lie zrowth factor-I action on hypothyroid rat pituitary cells: suppressi& of triiodothyronineinduced growth hormone secretion and messenger ribonucleic acid levels. Endocrinology 118~1483-1490 Cheung PT, Smith EP, Shimasaki S, Ling N, Chernausek SD 1991 Characterization of an insulin-like grosh factor binding protein (IGFBP-4) nroduced bv B104 rat neuronal cell line: chemical and biological properties and differential synthesis by sublines. Endocrinology 129:1006-10016 Conover CA, Ronk M, Lombana F, Powell DR 1990 Structural and biological characterization of bovine insulin-like growth factor binding protein-3. Endocrinology 1222795-2803 Hollander P. Stanlev F. Samuels HH. Thvroid and alucocorticoid hormones induce in&n receptors and inhibit insuli~ degradation in cultured GH cells. 61st Annual Meeting of The Endocrine Society, Anaheim, CA, 1979 (Abstract 497) Katakami H, Downs TR, Frohman LA 1986 Decreased hypothalamic growth hormone-releasing hormone content and pituitary responsiveness in hypothyroidism. J Clin Invest 77~1704-1711

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 January 2017. at 06:29 For personal use only. No other uses without permission. . All rights reserved.