00l3-7227/90/1266-3139$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society
Vol. 126, No. 6 Printed in U.S.A.
Regulation of Insulin-Like Growth Factor (IGF)-Binding Protein Synthesis by Insulin and IGF-I in Cultured Bovine Fibroblasts* CHERYL A. CONOVER Endocrine Research Unit, Mayo Clinic, Rochester, Minnesota 55905
ABSTRACT. Specific insulin-like growth factor-binding proteins (IGFBPs) are synthesized and secreted by bovine fibroblasts in vitro. By Western ligand blotting, three molecular forms of IGFBP were identified in conditioned medium from control cultures with mol wt (Mr) of 34,000, 28,000, and 24,000. Concentrations of these three IGFBP forms increased with time in serum-free conditioned medium without benefit of hormonal supplementation. Insulin and IGF-I were potent stimuli for IGFBP production by bovine fibroblasts, whereas bovine GH, epidermal growth factor, or steroid treatment had little or no effect. Insulin and IGF-I enhanced the production of 24,000,
T
HE INSULIN-like growth factors (IGFs) are GHdependent peptides structurally related to insulin that have anabolic and mitogenic effects in vivo and in vitro (1). IGFs have been purified from a variety of mammalian plasmas and have been found to have a high degree of sequence homology (2). A significant aspect of IGF physiology is that plasma and other biological fluids contain multiple proteins that specifically bind IGFs (3, 6). To date, three IGF-binding proteins (IGFBPs) have been purified and sequenced, IGFBP-1 (7), IGFBP-2 (8, 9), and IGFBP-3 (10),1 which have distinct molecular sizes, biochemical characteristics, and immunological determinants. Furthermore, IGFBPs may function in diverse ways to modulate the biological effects of the IGFs (11-14). Therefore, the IGFBPs are important determinants of IGF action, both quantitatively and qualitatively. Target cells for IGFs, including bone, liver, and muscle , cells, and fibroblasts, have been shown to produce different IGFBPs in vitro (15-18). The present study characterizes the IGFBP species produced by cultured bovine Received November 22,1989. Address requests for reprints to: Dr. Cheryl A. Conover, Endocrine Research Unit, Room 5-164 West Joseph Building, Mayo Clinic, Rochester, Minnesota 55905. * This work was supported by the Mayo Medical Foundation. 1 This nomenclature was discussed and agreed upon by consensus at the Workshop on IGF Binding Proteins, Vancouver, British Columbia, Canada, June 17-19,1989.
28,000, and 34,000 Mr IGFBPs in a dose-dependent fashion. Moreover, addition of low nanomolar concentrations of insulin or IGF-I to bovine fibroblast cultures specifically induced the secretion of a 42,000/38,000 Mr species of IGFBP, which corresponded in size to the IGF-binding subunit of the principal 150,000 Mr IGFBP complex in serum. After stimulation with insulin or IGF-I, bovine fibroblasts (3 x 105 cells) secreted approximately 30 ng/24 h 42,000/38,000 Mr IGFBP. Subunits of 42,000/38,000 Mr in bovine fibroblast-conditioned medium did not form macromolecular complexes in either the absence or presence of bovine GH. (Endocrinology 126: 3139-3145, 1990)
skin fibroblasts and demonstrates unique regulation of IGFBP synthesis by insulin and IGF-I in this cell model.
Materials and Methods Materials Crystalline bovine and human insulins were supplied by Eli Lilly Co. (Indianapolis, IN). Bovine GH (bGH) was obtained from the National Hormone and Pituitary Program (NIDDK, Baltimore, MD). IGF-I was purchased from Amgen Biologicals (Thousand Oaks, CA), and epidermal growth factor (EGF) from Collaborative Research, Inc. (Lexington, MA). Dexamethasone, progesterone, protamine sulfate, activated charcoal, and BSA were purchased from Sigma (St. Louis, MO). IGFBP purified from bovine serum was generously provided by Dr. David Powell (Baylor College of Medicine, Houston, TX). [126I] IGF-I and 14C-labeled rainbow mol wt (Mr) standards were obtained from Amersham (Arlington Heights, IL). Tissue culture media and medium supplements were purchased from Gibco (Grand Island, NY). Cell cultures Bovine dermal fibroblasts from a 3-yr-old male Hereford and a 2-yr-old female Holstein (GM06034 and GM06035, respectively), human dermal fibroblasts from a 24-yr-old male (GM03652A), and simian virus-40 (SV40)-transformed human fibroblasts (GM00637E) were purchased from the Human Genetic Mutant Cell Repository (Camden, NJ). Fibroblasts were cultured in Dulbecco's Modified Eagle's Medium (DMEM)
3139
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP
3140
supplemented with 100 U/ml penicillin, 100 Mg/ml streptomycin, and 4 mM glutamine, and containing 10% fetal bovine serum (HyClone Laboratories, Logan, UT). Cultures were used between passages 9-15 and were proven to be free of mycoplasma contamination. Conditioned medium Cells were detached by trypsinization and plated in 16-mm multiwells (Costar, Cambridge, MA). At confluency, fibroblasts were washed twice and preincubated in a 1:1 mixture (vol/vol) of Waymouth's medium-DMEM plus 0.1% BSA (SFM) for 24 h. The cells were then washed and the medium changed to SFM with or without experimental additions for the indicated times. At the end of the incubation period, the conditioned medium was centrifuged at 2000 X g at 4 C for 30 min and frozen at —20 C until the time of assay. Although assays were routinely performed within the week of the experiment, stability studies revealed no significant decrease in IGFBP activity in conditioned medium stored for up to 3 months. At the time of medium collection, cell counts were determined on triplicate wells with a Coulter counter (Coulter Electronics, Hialeah, FL). There was no significant change in cell number with experimental treatment under these culture conditions. Charcoal separation ligand binding assay IGF-binding activity in the medium was measured, and the assay method validated as previously described (18). Aliquots (5-20 /A) of conditioned medium, each dose in triplicate, were incubated with [125I]IGF-I overnight at 4 C. One percent activated charcoal containing 0.2 mg/ml protamine sulfate was added, and the samples were centrifuged at 4 C to separate bound from free IGF-I. Unconditioned medium was assayed in parallel and was subtracted from the total bound radioactivity to determine specific IGF-binding activity. Experimental results were compared at equivalent medium concentrations and only where dose-dependent relationships were linear.
Endo • 1990 Vol 126 • No 6
night at 4 C, and visualized by autoradiography, according to the method of Hossenlopp et al. (19). Scanning densitometry was performed using a UtroScan XL laser densitometer, and absorbance curves were integrated and compared using GelScan XL software (Pharmacia LKB Biotechnology). Different transfer efficiencies and binding affinities for the various IGFBPs preclude direct quantitative comparisons between IGFBP forms based upon band intensities and absorbance units. However, relative changes in band intensities for a particular IGFBP form can be evaluated using scanning densitometry.
Results Hormonal regulation of IGFBP secretion The effects of various hormones and growth factors on the secretion of IGFBPs by cultured bovine fibroblasts were investigated, and the results are presented in Table 1. Under basal serum-free conditions, there was very little IGFBP activity in 24-h conditioned medium. Treatment of bovine fibroblasts with bGH, progesterone, or EGF had no significant effect on IGFBP secretion. Dexamethasone had a small inhibitory effect (23% decrease) in one bovine cell line, but not in the other. In contrast, insulin (100 nM) increased IGFBP activity 3- to 4-fold in 24-h conditioned medium from both bovine fibroblast lines. Equivalent stimulatory effects were seen with bovine and human insulins. Acid chromatography of the various conditioned media, to remove potential interference from endogenous IGF peptides, had no effect on relative IGFBP activity (data not shown). To determine whether changes in IGFBP secretion TABLE 1. Hormonal control of bovine fibroblast IGFBP secretion IGFBP ACTIVITY (% of control)0 GM6034
Superose-12 gel chromatography Conditioned medium and serum samples (200 jul) were fractionated on a FPLC Superose-12 10/30 column (Pharmacia LKB Biotechnology, Inc., Piscataway, NJ) equilibrated in 50 mM Tris-HCl and 150 mM sodium chloride containing 0.5% BSA. Fractions (500 n\) were collected and assayed for IGFbinding activity as described in the preceding paragraph. The column was calibrated with [125I]IGF-I and with reference proteins of known Mr (0-amylase, 200,000; 7-globulin, 150,000; BSA, 66,000; chymotrypsinogen, 25,700; ribonuclease-a, 13,700; cytochrome-c, 12,500). Western ligand blots Unreduced conditioned medium and serum samples were electrophoresed through a sodium dodecyl sulfate-polyacrylamide gel using a 7.5-15% linear gradient, and separated proteins were electroblotted onto nitrocellulose filters (0.45-/^m pore size) using a BioTrans Unit (Gelman Sciences, Ann Arbor, MI). Filters were blocked, labeled with [125I]IGF-I over-
bGH DEX P EGF INS
99 ± 77 ± 100 ± 115 ± 298 ±
5 (7) 5 (6)° 1 (3) 10 (4) 14 (9)"
GM6035 108 ± 8 (2) 92 ± 4 (3) 96 ± 4 (2) 106 ± 6 (3) 389 ± 14 (3)°
Confluent bovine fibroblast cultures were washed and preincubated for 24 h in SFM, and the medium was changed to SFM with or without experimental additions. Conditioned medium was collected after 24 h. bGH, (1 Mg/ml); DEX, dexamethasone (100 nM); P, progesterone (1 ixM); EGF, 100 ng/ml; INS, insulin (100 nM). Significance was determined by Student's t test. " IGFBP activity in the conditioned medium from bovine fibroblast lines GM6034 and GM6035 was determined by charcoal separation ligand binding, as described in Materials and Methods. Results (mean ± SE of the number of experiments shown in parentheses) are presented as a percentage of the IGFBP activity in control (untreated) cultures. Conditioned medium from control cultures bound 6 ± 0.6% and 12 ± 0.9% [125I]IGF-I at 5 and 10 p\, respectively. Medium samples from insulin-treated cultures were diluted when necessary for comparisons in the linear range of the assay. 0 P < 0.01 vs. control. * P < 0.001 vs. control.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP
were due to the release of stored IGFBP or to the production and release of newly synthesized IGFBP, the effect of a protein synthesis inhibitor was examined. When cycloheximide (1 ng/m\) was added to the cell cultures, there was no apparent IGFBP activity in 24-h conditioned medium in either the absence or presence of insulin (Table 2). The cells retained viability, since replacement of the medium with fresh SFM resulted in complete recovery of basal and insulin-stimulated IGFBP production. Therefore, bovine fibroblasts appear to actively synthesize and secrete IGFBPs, and this synthesis is regulated by insulin.
B
30-
21.5Serum
Molecular forms of bovine fibroblast IGFBP
To identify the particular IGFBP species secreted by bovine fibroblasts, Western ligand blotting was performed. Figure 1 shows the migration of IGFBPs from conditioned medium of cultured cells and from serum after sodium dodecyl sulfate-poly aery lamide gel electrophoresis (nonreducing conditions), transfer onto nitrocellulose, and incubation with [125I]IGF-I. In bovine and human serum, multiple bands were observed, two major bands of 42,000 and 38,000 Mr and additional bands at 34,000, 31,000, 28,000, and 24,000 Mr, similar to previously reported Western ligand blot patterns of mammalian serum and plasma (3, 4, 20). The 42,000 and 38,000 Mr bands have been shown to represent different glycoslation states of the binding subunit of the 150,000 Mr major plasma complex (3, 4). Conditioned medium from normal adult human fibroblasts contained the IGFBP forms seen in serum, as has been reported previously (18). Conditioned medium from SV40-transformed human fibroblasts also contained 42,000/38,000, 28,000, TABLE 2. Effect of cycloheximide on bovine fibroblast IGFBP production % of [125I]IGF-I bound" Treatment -INS +INS Recovery -INS +INS
Control
CX
15 30
5 7
10 29
11 27
3141
HF CM
I I SV40 HF CM
FIG. 1. Western ligand blot of IGFBPs. Serum and conditioned medium samples were separated on a 7.5-15% gradient sodium dodecyl sulfate-polyacrylamide gel under nonreducing conditions, electroblotted onto nitrocellulose filters, and after incubation with [126I]IGF-I, identified by autoradiography. +, Treatment with 100 nM insulin. BF CM, Bovine fibroblast 72-h conditioned medium; Serum B, Bovine; H, human; HF CM, human fibroblast 72-h conditioned medium; SV40 HF CM, SV40-transformed human fibroblast 48-h conditioned medium.
and 24,000 Mr IGFBPs, with the 24,000 Mr form being predominant. Bovine fibroblasts secreted low levels of 24,000, 28,000, and 34,000 Mr IGFBPs under control conditions (Fig. 1). In general, the 24,000 Mr species was the major IGFBP in the unstimulated state. Treatment of bovine fibroblast cultures with 100 nM insulin resulted in a relative 2- to
4-fold increase in these three IGFBP forms in the conditioned medium. Moreover, in the presence of insulin, bovine fibroblasts secreted significant amounts of 42,000 and 38,000 Mr IGFBP subunits, which were undetectable under basal conditions. In comparison, addition of 100 nM insulin to normal adult human fibroblasts or SV40transformed human fibroblasts resulted in little or no apparent change in the intensity or pattern of secreted IGFBPs. Regulation of IGFBP by insulin and IGF-I
Bovine fibroblasts were washed, preincubated in SFM for 24 h, and changed to SFM with or without insulin (1 tig/ml) in the presence or absence of cycloheximide (CX; 1 /ig/ml). After a 24-h treatment period, conditioned medium was collected and the cells were washed twice and incubated in SFM for a 6-h washout period. Medium was then replaced with SFM with or without insulin, and the conditioned medium was collected after a 24-h recovery period. 0 Results are expressed as a percentage of the total [126I]IGF-I added that is specifically bound. Values for 10 /xl conditioned medium samples are presented.
Figure 2 presents a time course for IGFBP production2 by bovine fibroblasts. Over 96 h in serum-free culture, bovine fibroblasts constitutively secreted 24,000, 28,000, and 34,000 Mr IGFBP forms. Treatment with insulin (100 nM) increased the production of all three forms at every time point, with a maximal effect seen by 72 h. Densitometic analysis of the band intensities of two separate experiments indicated that the 24,000, 28,000, and 34,000 Mr bands were increased 1.6- and 3.4-fold, 3.1- and 4.2-fold, and 1.8- and 24-fold, respectively, by 2
In this paper, production is used to denote a net effect, since potential degradation of IGFBPs was not determined.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
3142
REGULATION OF BOVINE FIBROBLAST IGFBP
Mr x10~ 3
Endo • 1990 Vol 126 • No 6
Mrx10-3
46-
46-
30-
3021.5-
21.5-
24h
48h
72h
96h
96h
INS (nM)
IGF-I (nM)
FIG. 2. Western ligand blot: time course of insulin (INS) and IGF-I stimulation. Conditioned medium samples from bovine fibroblasts incubated for the indicated times without or with 100 nM insulin or 10 nM IGF-I were analyzed by Western ligand blotting, as described in Fig. 1.
FIG. 3. Western ligand blot: effects of insulin (INS) and IGF-I. Conditioned medium from bovine fibroblast cultures untreated (C, control) or treated with various concentrations of insulin or IGF-I for 24 h were analyzed by Western ligand blotting, as described in Fig. 1.
72 h of insulin treatment. Moreover, insulin treatment for 24 h induced the apperance of a 42,000/38,000 Mr IGFBP form, which increased progressively over 96 h. There was no 42,000/38,000 Mr doublet evident in control culture medium, even with a 96-h incubation. IGF-I also increased the production of 24,000, 28,000, and 34,000 Mr IGFBPs and induced the secretion of 42,000/37,000 Mr IGFBP by bovine fibroblasts (Fig. 2B). The effects of IGF-I (10 nM) were equivalent to those of insulin (100 nM), except that an additional 31,000 Mr IGFBP band was sometimes observed with IGF-I treatment. Figure 2B presents a Western ligand blot of 96-h conditioned medium, but the effects of insulin and IGFI were also similar at 24, 48, and 72 h (data not shown). When [125I]IGF-II was employed as the radioligand, the only observed change was a relative 2-fold increase in the intensity of the 42,000/38,000 Mr bands from conditioned medium samples of insulin- and IGF-Itreated bovine fibroblasts and serum, indicating a slight preference for IGF-II by the 42,000/38,000 Mr IGFBP form, as has been reported previously for IGFBP-3 (21). The 24,000, 28,000, and 34,000 Mr IGFBP species appeared to bind [125I]IGF-I and [125I]IGF-II equivalent^ (data not shown). By Western ligand blot analysis, the dose-response effects of insulin and IGF-I on the regulation of bovine fibroblast IGFBP production were compared. Conditioned medium was collected 24 or 48 h after the addition of various concentrations of the two peptides, and the results of one of three experiments are presented in Fig. 3. Insulin and IGF-I, at low concentrations, were able to augment synthesis of the 24,000, 28,000, and 34,000 Mr forms of IGFBPs and induce secretion of the 42,000/ 38,000 Mr species. A stimulatory effect on all IGFBP forms was seen at 0.01 nM insulin. A maximal effect on
24,000, 28,000, and 34,000 Mr IGFBP production was observed with 1-10 nM insulin. IGF-I showed similar dose-response characteristics, except the magnitude of the response was less obvious at 0.01 and 0.1 nM IGF-I. A distinct stimulatory effect of 0.01 nM IGF-I was seen in one of the three experiments. In all three experiments, insulin and IGF-I at 1 nM had near-equivalent effects on IGFBP production by bovine fibroblasts. Furthermore, insulin and IGF-I treatment resulted in striking dosedependent increases in 42,000/38,000 Mr IGFBP secretion. Similar dose-response effects of insulin and IGF-I on IGFBP production were seen with a 48-h incubation period (data not shown). IGF-I and insulin appear to specifically affect IGFBP production independent of changes in general protein synthesis. Whereas IGF-I and insulin at 1 nM have similar near-maximal effects on IGFBP secretion, half-maximal effects on aminoisobutyric acid uptake in bovine fibroblasts were found with 2 and 20 nM IGF-I and insulin, respectively (data not shown). Quantitation of 42,000/38,000 Mr IGFBP production IGFBP induction by insulin and IGF-I in bovine fibroblast cultures was quantitated using purified bovine 42,000/38,000 Mr IGFBP (>98% pure) and Western ligand blotting. Conditioned medium samples from bovine fibroblasts treated with 10 nM insulin or IGF-I were run in parallel with known concentrations (5-160 ng/ml) of pure bovine 42,000/38,000 Mr IGFBP. Data were obtained by densitometric scanning of films from different exposure times, and linear dependence of optical density on the amount of IGFBP in the sample was validated. By densitometric analyses of band intensities and construction of standard curves for three separate experiments, it could be estimated that confluent cultures of
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP
3143
bovine fibroblasts (3.0 ± 0.3 x 105 cells) secrete 32 ± 2 and 36 ± 7 ng/ml 42,000/38,000 Mr IGFBP with 24 h of insulin or IGF-I stimulation, and 87 ± 14 and 124 ± 53 ng/ml after 96 h of insulin or IGF-I stimulation, respectively. Bovine serum contained 6.3 ± 0.6 /ug/ml and human serum contained 8.2 ± 0.8 /xg/ml 42,000/38,000 Mr IGFBP. Serum concentrations may be underestimated because of the labeling intensities of the bands.
matography fractions, 2) Western ligand blot analysis of neutral chromatography fractions, and 3) preincubation of conditioned medium with [125I]IGF-I before neutral chromatography. Parallel analyses of serum samples demonstrated 150,000 Mr IGFBP complexes under these conditions (data not shown).
Neutral chromatography of bovine fibroblast-conditioned medium
Bovine fibroblasts in culture synthesize and secrete low Mr IGFBPs which are precisely controlled by insulin and IGF-I in a time- and dose-dependent manner. Under basal serum-free conditions, bovine fibroblasts produce IGFBPs with Mr of 24,000, 28,000, and 34,000. The 24,000 Mr IGFBP was a predominant secretory product of bovine fibroblasts, as has been shown for normal adult human fibroblasts (18), transformed human fibroblasts, L6 rat muscle cells (17), and human breast cancer cells (22). This IGFBP form has not been sequenced as yet, but appears to be distinct from IGFBP1, IGFBP-2, and IGFBP-3 (18, 22, 23). The 34,000 Mr form is similar in size to the binding protein designated IGFBP-2, purified from the conditioned medium of MDBK and BRL3A cell cultures (8, 9, 24). This 34,000 Mr form secreted by cultured bovine fibroblasts did not exhibit a binding preference for IGF-II over IGF-I, which is characteristic of IGFBP-2 (8, 24), although relative affinities for the two peptides were not directly compared. The 28,000 Mr form is the same size as IGFBP-1 secreted by HepG2 cells. However, the secretion of bovine fibroblast 28,000 Mr IGFBP was stimulated by insulin, whereas insulin inhibits HepG2 IGFBP-1 production (18, 25). The 28,000 Mr form in human fibroblast-conditioned medium was also previously shown to be distinct from IGFBP-1 by immunological and physiological criteria (18). Whether these 34,000 and 28,000 Mr IGFBPs are additional IGFBP species or posttranslational modifications of one of the other forms remains to be determined. Production of the 34,000,28,000, and 24,000 Mr IGFBP forms by bovine fibroblasts was positively modulated by both insulin and IGF-I at subnanomolar concentrations. McCusker and Clemmons (17) similarly reported that insulin and IGF-I increased the secretion of 31,000 and 24,000 MJGFBPs in rat 1-6 muscle cells. In comparison, we have found the opposite effect of insulin in HepG2 cells, where insulin markedly and rapidly suppressed IGFBP-1 synthesis (18, 25). Likewise, IGFBP-1 concentrations in vivo are inversely correlated with circulating insulin concentrations (26). Clearly, insulin and IGF-I have differential effects on the various IGFBP species.
Conditioned medium samples from bovine fibroblasts stimulated by insulin (100 nM) and bGH (1 ng/m\), individually or in combination, were subjected to neutral chromatography on a Superose-12 column. Profiles of IGFBP activity recovered after fractionation are shown in Fig. 4. In medium from insulin-stimulated cells, a single peak of IGFBP activity eluted between albumin and ribonuclease-a markers (approximate Mr = 20,00050,000). Conditioned medium from bovine fibroblasts stimulated by insulin in the presence of bGH had an elution profile identical to that of insulin stimulation alone, with a single peak of IGFBP activity at 20,00050,000 Mr. There was no evidence of IGFBP activity in fractions corresponding to Mr greater than 60,000. In contrast, chromatographed human serum had two peaks of IGFBP activity, a high Mr peak eluting between /?amylase and albumin (approximate Mr = 150,000) and a 20,000-50,000 Mr peak. Conditioned medium from bovine fibroblasts treated for 72 h with IGF-I (10 nM) and bGH (1 Mg/ml) was also chromatographed under neutral conditions, as described above. No IGF-binding activity could be detected in high Mr fractions, as assessed by 1) assay of IGF-binding activity after acid chromatography of the neutral chro70 i-
60 T3
c o (D
50 40 30 20 10
0110
15
20
25
30
35
40
Fraction number FIG. 4. Superose-12 chromatography at neutral pH. Conditioned medium from bovine fibroblasts treated with 100 nM insulin (O) or insulin plus 1 fig/ml GH (A) and human serum samples (•) were gel filtered on a Superose-12 column in 50 mM Tris, 150 mM NaCl, and 0.5% BSA, pH 7.4. Fractions were assayed for IGFBP activity by charcoal separation ligand binding. The arrows indicate the elution volume of proteins used as Mr markers (xiO3). Vo, Void volume.
Discussion
Of particular interest, insulin and IGF-I specifically and markedly induced the secretion of the 42,000/38,000 Mr form of IGFBP, which corresponds in size to the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
3144
REGULATION OF BOVINE FIBROBLAST IGFBP
major IGFBP species in mammalian serum. This IGFBP is analogous to IGFBP-3, the acid-stable binding subunit of the 150,000 Mr complex in human plasma (10, 21). Confluent cultures of bovine fibroblasts secrete approximately 30 ng/24 h 42,000/38,000 Mr IGFBP after insulin or IGF-I treatment, as determined from a standard curve using known quantities of purified bovine 42,000/38,000 Mr IGFBP. With this method, bovine and human serum was estimated to contain approximately 6-8 ixg/m\ 42,000/38,000 Mr IGFBP, which agrees with measurements of this IGFBP form in human plasma at 6-9 ng/ ml using a specific RIA (27). Whether in vivo this form acts primarily at the site of synthesis to modulate IGF action or contributes to circulating IGFBP concentrations remains to be investigated. In human plasma, levels of the 42,000/38,000 Mr IGFBP species reflect GH status (27). The 42,000/38,000 Mr IGFBP secreted by cultured human fibroblasts was also found to be GH dependent (18). In contrast, GH had no effect on IGFBP production in the bovine fibroblast system. Instead, insulin and IGF-I potently and specifically induced the synthesis of 42,000/38,000 Mr IGFBP by bovine fibroblasts. Species differences may account for differential regulatory mechanisms for 42,000/38,000 Mr IGFBP in the human and bovine systems. However, bovine IGF-I has been sequenced and found to be identical to human IGF-I (28). Therefore, bovine and human IGFBP forms and their control mechanisms would be expected to be highly homologous, if not identical. Alternatively, IGF-I may be the principal regulator of the 42,000/38,000 Mr class of IGFBPs, and GH could have indirect effects. In plasma, IGF-I concentrations, which are GH dependent (2), could be responsible for the observed GH dependency of 42,000/38,000 Mr IGFBP. Similarily, human fibroblasts have been shown to secrete IGF-like peptides, the synthesis of which can be stimulated with GH treatment (29-31). This could account for 42,000/38,000 Mr IGFBP production by human fibroblasts in the basal state and for GHdependent increases in IGFBP synthesis (18). It is of note that cultured bovine fibroblasts do not appear to secrete IGF peptide (my unpublished observations). Consistent with this hypothesis, Zapf et at. (32) found that IGF-I induced 42,000-49,000 Mr IGFBP in plasma of hyppophysectomized and diabetic rats. Furthermore, the ontological appearance of 43,000/39,000 Mr IGFBP subunits corresponded in time to the appearance of IGF-I in fetal porcine serum (20), and a correlation between the 42,000-49,000 Mr cluster and IGF-I status was observed in rat osteoblastic cells in vitro (15). In addition to IGF-I, insulin appears to play an important role in regulating IGFBP production, as seen in the bovine fibroblast model system, and, thus, may coordinate IGF bioactivity with nutritional status.
Endo • 1990 Vol 126 • No 6
Although GH may not be the primary regulator of 42,000/38,000 Mr IGFBP synthesis, the findings of Zapf et al. (32) indicated that GH was necessary for association of these binding subunits into the 150,000 Mr circulating IGFBP complex. This theoretically could come about by GH-regulated production of the acid-labile subunit, which, along with IGF peptide, is necessary for complex formation (33). Although bovine fibroblasts secrete large amounts of 42,000/38,000 Mr IGF-binding subunits in response to insulin and IGF-I, in the presence of GH these binding subunits did not form any large Mr IGFBP complex, as could be found in serum. Therefore, bovine fibroblasts in culture may not be responsive to GH or may be unable to produce the acid-labile subunit necessary for complete association. Thus, physiological concentrations of insulin and IGFI stimulate IGFBP production in cultured bovine fibroblasts. The mechanisms by which the different IGFBP forms affect IGF endocrine and autocrine/paracrine functions are unknown; however, both inhibiting and potentiating effects have been noted in vitro (11-14). The distribution of each IGFBP form and the hormonal and nutritional control of their synthesis and secretion may be important determinants of IGF action in normal and abnormal physiology.
Acknowledgments I would like to thank Laurie Bale and Jay Clarkson for their excellent technical assistance.
References 1. Froesch ER, Schmid C, Schwander J, Zapf JA 1985 Actions of insulin-like growth factors. Rev Physiol 47:443 2. Daughaday WH, Rotwein P 1989 Insulin-like growth factors I and II. Peptide, messenger ribonucleic acid and gene structures, serum, and tissue concentrations. Endocr Rev 10:68 3. Hardouin S, Hossenlopp P, Segovia B, Seurin D, Portolan G, Lassare C, Binoux M 1987 Heterogeneity of insulin-like growth factor binding proteins and relationships between structure and affinity. I. Circulating forms in man. Eur J Biochem 170:121 4. Yang YW-H, Wang J-F, Orlowski CC, Nissley SP, Rechler MM 1989 Structure, specificity, and regulation of the insulin-like growth factor-binding proteins in adult rat serum. Endocrinology 125:1540 5. Povoa G, Enberg G, Jornvall H, Hall K 1984 Isolation and characterization of a somatomedin binding protein from mid-term amniotic fluid. Eur J Biochem 144:199 6. Hossenlopp P, Seurin D, Segovia-Quinson B, Binoux M 1986 Identification of an insulin-like growth factor binding protein in human cerebrospinal fluid with a selective affinity for IGF-II. FEBS Lett 208:439 7. Lee Y-L, Hintz RL, James PM, Lee PDK, Shively JE, Powell DR 1988 Insulin-like growth factor (IGF) binding protein complementary deoxyribonucleic acid from human HepG2 hepatoma cells: predicted protein sequence suggests an IGF binding domain different from those of the IGF-I and IGF-II receptors. Mol Endocrinol 2:404 8. Binkert C, Landwehr J, Mary J-L, Schwander J, Heinrich G 1989 Cloning, sequence analysis and expression of a cDNA encoding a novel insulin-like growth factor binding protein (IGFBP-2). EMB0
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP J 8:2497 9. Brown AL, Chiariotti L, Orlowski CC, Mehlman T, Burgess WH, Ackerman EJ, Bruni CB, Rechler MM 1989 Nucleotide sequence and expression of a cDNA clone encoding a fetal rat binding protein for insulin-like growth factors. J Biol Chem 264:5148 10. Wood WI, Cachianes G, Henzel WJ, Winslow GA, Spencer SA, Hellmiss R, Martin J, Baxter RC 1988 Cloning and expression of the growth hormone-dependent insulin-like growth factor-binding protein. Mol Endocrinol 2:1176 11. Ritvos 0, Ranta T, Jalkanen J, Suikkari A-M, Voutilainen R, Bohn H, Rutanen E-M 1988 Insulin-like growth factor (IGF) binding protein from human decidua inhibits the binding and biological action of IGF-I in cultured choriocarcinoma cells. Endocrinology 122:2150 12. Busby WH, Klapper DG, Clemmons DR 1988 Purification of a 31,000-dalton insulin-like growth factor binding protein from human amniotic fluid: isolation of two forms with different biologic actions. J Biol Chem 263:14203 13. Ross M, Francis GL, Szabo L, Wallace JC, Ballard JF 1989 Insulinlike growth factor (IGF) binding proteins inhibit the biological activities of IGF-1 and IGF-2 but not des-(l-3)-IGF-l. Biochem J 258:267 14. DeMellow JSM, Baxter RC 1988 Growth hormone-dependent insulin-like growth factor (IGF) binding protein both inhibits and potentiates IGF-I-stimulated DNA synthesis in human skin fibroblasts. Biochem Biophys Res Commun 156:199 15. Schmid C, Zapf J, Froesch ER 1989 Production of carrier proteins for insulin-like growth factors (IGFs) by rat osteoblastic cells. FEBS Lett 244:328 16. Moses AC, Freinkel AJ, Knowles BB, Aden DP 1983 Demonstration that a human hepatoma cell line produces a specific insulinlike growth factor carrier protein. J Clin Endocrinol Metab 56:1003 17. McCusker RH, Clemmons DR 1988 Insulin-like growth factor binding protein secretion by muscle cells: effect of cellular differention and proliferation. J Cell Physiol 137:505 18. Conover CA, Liu F, Powell D, Rosenfeld RG, Hintz RL 1989 Insulin-like growth factor binding proteins from cultured human fibroblasts: characterization and hormonal regulation. J Clin Invest 83:852 19. Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M 1986 Analysis of serum insulin-like growth factor binding proteins using Western blotting: use of the method for titration of the binding proteins and competitive binding studies. Anal Biochem 154:138 20. McCusker RH, Campion DR, Jones WK, Clemmons DR 1989 The
21. 22. 23.
24.
25. 26.
27. 28. 29. 30. 31. 32.
33.
3145
insulin-like growth factor binding proteins of porcine serum: endocrine and nutritional regulation. Endocrinology 125:501 Martin JL, Baxter RC 1986 Insulin-like growth factor-binding protein from human plasma: purification and characterization. J Biol Chem 261:8754 DeLeon DD, Wilson DM, Bakker B, Lamson G, Hintz RL, Rosenfeld RG 1989 Characterization of insulin-like growth factor binding proteins from human breast cancer cells. Mol Endocrinol 3:567 Rosenfeld RG, Pham H, Oh Y, Lamson G, Guidice LC 1990 Identification of insulin-like growth factor binding protein-2 (IGFBP-2) and a low molecular weight IGFBP in human seminal plasma. J Clin Endocrinol Metab 70:551 Szabo L, Mottershead DG, Ballard JF, Wallace JC 1988 The bovine insulin-like growth factor (IGF) binding protein purified from conditioned medium requires the N-terminal tripeptide in IGF-I for binding. Biochem Biophys Res Commun 151:207 Conover CA, Lee PDK 1990 Insulin regulation of IGF binding protein production in cultured HepG2 cells. J Clin Endocrinol Metab 70:1062 Suikkari AM, Koivisto VA, Rutanen EM, Yki-Jarvinen H, Karonen SL, Seppala M 1988 Insulin regulates the serum levels of low molecular weight insulin-like growth factor-binding protein. J Clin Endocrinol Metab 66:266 Baxter RC, Martin JL 1986 Radioimmunoassay of growth hormone-dependent insulinlike growth factor binding protein in human plasma. J Clin Invest 78:1504 Honegger A, Humbel RE 1986 Insulin-like growth factors I and II in fetal and adult bovine serum. J Biol Chem 261:569 Atkison PR, Weidman ER, Blaimick B, Bala RM 1980 Release of somatomedin-like activity by cultured WI-38 human fibroblasts. Endocrinology 106:2006 Clemmons DR, Underwood LE, Van Wyk JJ 1981 Hormonal control of immunoreactive somatomedin production by cultured human fibroblasts. J Clin Invest 67:10 Conover CA, Baker BK, Hintz RL 1989 Cultured human fibroblasts secrete insulin-like growth factor IA prohormone. J Clin Endocrinol Metab 69:25 Zapf J, Hauri C, Waldvogel M, Futo E, Hasler H, Binz K, Guler HP, Schmid C, Froesch ER 1989 Recombinant human insulin-like growth factor I induces its own specific carrier protein in hypophysectomized and diabetic rats. Proc Natl Acad Sci USA 86:3813 Baxter RC, Martin JL, Beniac VA 1989 High molecular weight insulin-like growth factor binding protein complex: purification and properties of the acid-labile subunit from human serum. J Biol Chem 264:11843
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
Vol. 126, No. 6 Printed in U.S.A.
Regulation of Insulin-Like Growth Factor (IGF)-Binding Protein Synthesis by Insulin and IGF-I in Cultured Bovine Fibroblasts* CHERYL A. CONOVER Endocrine Research Unit, Mayo Clinic, Rochester, Minnesota 55905
ABSTRACT. Specific insulin-like growth factor-binding proteins (IGFBPs) are synthesized and secreted by bovine fibroblasts in vitro. By Western ligand blotting, three molecular forms of IGFBP were identified in conditioned medium from control cultures with mol wt (Mr) of 34,000, 28,000, and 24,000. Concentrations of these three IGFBP forms increased with time in serum-free conditioned medium without benefit of hormonal supplementation. Insulin and IGF-I were potent stimuli for IGFBP production by bovine fibroblasts, whereas bovine GH, epidermal growth factor, or steroid treatment had little or no effect. Insulin and IGF-I enhanced the production of 24,000,
T
HE INSULIN-like growth factors (IGFs) are GHdependent peptides structurally related to insulin that have anabolic and mitogenic effects in vivo and in vitro (1). IGFs have been purified from a variety of mammalian plasmas and have been found to have a high degree of sequence homology (2). A significant aspect of IGF physiology is that plasma and other biological fluids contain multiple proteins that specifically bind IGFs (3, 6). To date, three IGF-binding proteins (IGFBPs) have been purified and sequenced, IGFBP-1 (7), IGFBP-2 (8, 9), and IGFBP-3 (10),1 which have distinct molecular sizes, biochemical characteristics, and immunological determinants. Furthermore, IGFBPs may function in diverse ways to modulate the biological effects of the IGFs (11-14). Therefore, the IGFBPs are important determinants of IGF action, both quantitatively and qualitatively. Target cells for IGFs, including bone, liver, and muscle , cells, and fibroblasts, have been shown to produce different IGFBPs in vitro (15-18). The present study characterizes the IGFBP species produced by cultured bovine Received November 22,1989. Address requests for reprints to: Dr. Cheryl A. Conover, Endocrine Research Unit, Room 5-164 West Joseph Building, Mayo Clinic, Rochester, Minnesota 55905. * This work was supported by the Mayo Medical Foundation. 1 This nomenclature was discussed and agreed upon by consensus at the Workshop on IGF Binding Proteins, Vancouver, British Columbia, Canada, June 17-19,1989.
28,000, and 34,000 Mr IGFBPs in a dose-dependent fashion. Moreover, addition of low nanomolar concentrations of insulin or IGF-I to bovine fibroblast cultures specifically induced the secretion of a 42,000/38,000 Mr species of IGFBP, which corresponded in size to the IGF-binding subunit of the principal 150,000 Mr IGFBP complex in serum. After stimulation with insulin or IGF-I, bovine fibroblasts (3 x 105 cells) secreted approximately 30 ng/24 h 42,000/38,000 Mr IGFBP. Subunits of 42,000/38,000 Mr in bovine fibroblast-conditioned medium did not form macromolecular complexes in either the absence or presence of bovine GH. (Endocrinology 126: 3139-3145, 1990)
skin fibroblasts and demonstrates unique regulation of IGFBP synthesis by insulin and IGF-I in this cell model.
Materials and Methods Materials Crystalline bovine and human insulins were supplied by Eli Lilly Co. (Indianapolis, IN). Bovine GH (bGH) was obtained from the National Hormone and Pituitary Program (NIDDK, Baltimore, MD). IGF-I was purchased from Amgen Biologicals (Thousand Oaks, CA), and epidermal growth factor (EGF) from Collaborative Research, Inc. (Lexington, MA). Dexamethasone, progesterone, protamine sulfate, activated charcoal, and BSA were purchased from Sigma (St. Louis, MO). IGFBP purified from bovine serum was generously provided by Dr. David Powell (Baylor College of Medicine, Houston, TX). [126I] IGF-I and 14C-labeled rainbow mol wt (Mr) standards were obtained from Amersham (Arlington Heights, IL). Tissue culture media and medium supplements were purchased from Gibco (Grand Island, NY). Cell cultures Bovine dermal fibroblasts from a 3-yr-old male Hereford and a 2-yr-old female Holstein (GM06034 and GM06035, respectively), human dermal fibroblasts from a 24-yr-old male (GM03652A), and simian virus-40 (SV40)-transformed human fibroblasts (GM00637E) were purchased from the Human Genetic Mutant Cell Repository (Camden, NJ). Fibroblasts were cultured in Dulbecco's Modified Eagle's Medium (DMEM)
3139
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP
3140
supplemented with 100 U/ml penicillin, 100 Mg/ml streptomycin, and 4 mM glutamine, and containing 10% fetal bovine serum (HyClone Laboratories, Logan, UT). Cultures were used between passages 9-15 and were proven to be free of mycoplasma contamination. Conditioned medium Cells were detached by trypsinization and plated in 16-mm multiwells (Costar, Cambridge, MA). At confluency, fibroblasts were washed twice and preincubated in a 1:1 mixture (vol/vol) of Waymouth's medium-DMEM plus 0.1% BSA (SFM) for 24 h. The cells were then washed and the medium changed to SFM with or without experimental additions for the indicated times. At the end of the incubation period, the conditioned medium was centrifuged at 2000 X g at 4 C for 30 min and frozen at —20 C until the time of assay. Although assays were routinely performed within the week of the experiment, stability studies revealed no significant decrease in IGFBP activity in conditioned medium stored for up to 3 months. At the time of medium collection, cell counts were determined on triplicate wells with a Coulter counter (Coulter Electronics, Hialeah, FL). There was no significant change in cell number with experimental treatment under these culture conditions. Charcoal separation ligand binding assay IGF-binding activity in the medium was measured, and the assay method validated as previously described (18). Aliquots (5-20 /A) of conditioned medium, each dose in triplicate, were incubated with [125I]IGF-I overnight at 4 C. One percent activated charcoal containing 0.2 mg/ml protamine sulfate was added, and the samples were centrifuged at 4 C to separate bound from free IGF-I. Unconditioned medium was assayed in parallel and was subtracted from the total bound radioactivity to determine specific IGF-binding activity. Experimental results were compared at equivalent medium concentrations and only where dose-dependent relationships were linear.
Endo • 1990 Vol 126 • No 6
night at 4 C, and visualized by autoradiography, according to the method of Hossenlopp et al. (19). Scanning densitometry was performed using a UtroScan XL laser densitometer, and absorbance curves were integrated and compared using GelScan XL software (Pharmacia LKB Biotechnology). Different transfer efficiencies and binding affinities for the various IGFBPs preclude direct quantitative comparisons between IGFBP forms based upon band intensities and absorbance units. However, relative changes in band intensities for a particular IGFBP form can be evaluated using scanning densitometry.
Results Hormonal regulation of IGFBP secretion The effects of various hormones and growth factors on the secretion of IGFBPs by cultured bovine fibroblasts were investigated, and the results are presented in Table 1. Under basal serum-free conditions, there was very little IGFBP activity in 24-h conditioned medium. Treatment of bovine fibroblasts with bGH, progesterone, or EGF had no significant effect on IGFBP secretion. Dexamethasone had a small inhibitory effect (23% decrease) in one bovine cell line, but not in the other. In contrast, insulin (100 nM) increased IGFBP activity 3- to 4-fold in 24-h conditioned medium from both bovine fibroblast lines. Equivalent stimulatory effects were seen with bovine and human insulins. Acid chromatography of the various conditioned media, to remove potential interference from endogenous IGF peptides, had no effect on relative IGFBP activity (data not shown). To determine whether changes in IGFBP secretion TABLE 1. Hormonal control of bovine fibroblast IGFBP secretion IGFBP ACTIVITY (% of control)0 GM6034
Superose-12 gel chromatography Conditioned medium and serum samples (200 jul) were fractionated on a FPLC Superose-12 10/30 column (Pharmacia LKB Biotechnology, Inc., Piscataway, NJ) equilibrated in 50 mM Tris-HCl and 150 mM sodium chloride containing 0.5% BSA. Fractions (500 n\) were collected and assayed for IGFbinding activity as described in the preceding paragraph. The column was calibrated with [125I]IGF-I and with reference proteins of known Mr (0-amylase, 200,000; 7-globulin, 150,000; BSA, 66,000; chymotrypsinogen, 25,700; ribonuclease-a, 13,700; cytochrome-c, 12,500). Western ligand blots Unreduced conditioned medium and serum samples were electrophoresed through a sodium dodecyl sulfate-polyacrylamide gel using a 7.5-15% linear gradient, and separated proteins were electroblotted onto nitrocellulose filters (0.45-/^m pore size) using a BioTrans Unit (Gelman Sciences, Ann Arbor, MI). Filters were blocked, labeled with [125I]IGF-I over-
bGH DEX P EGF INS
99 ± 77 ± 100 ± 115 ± 298 ±
5 (7) 5 (6)° 1 (3) 10 (4) 14 (9)"
GM6035 108 ± 8 (2) 92 ± 4 (3) 96 ± 4 (2) 106 ± 6 (3) 389 ± 14 (3)°
Confluent bovine fibroblast cultures were washed and preincubated for 24 h in SFM, and the medium was changed to SFM with or without experimental additions. Conditioned medium was collected after 24 h. bGH, (1 Mg/ml); DEX, dexamethasone (100 nM); P, progesterone (1 ixM); EGF, 100 ng/ml; INS, insulin (100 nM). Significance was determined by Student's t test. " IGFBP activity in the conditioned medium from bovine fibroblast lines GM6034 and GM6035 was determined by charcoal separation ligand binding, as described in Materials and Methods. Results (mean ± SE of the number of experiments shown in parentheses) are presented as a percentage of the IGFBP activity in control (untreated) cultures. Conditioned medium from control cultures bound 6 ± 0.6% and 12 ± 0.9% [125I]IGF-I at 5 and 10 p\, respectively. Medium samples from insulin-treated cultures were diluted when necessary for comparisons in the linear range of the assay. 0 P < 0.01 vs. control. * P < 0.001 vs. control.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP
were due to the release of stored IGFBP or to the production and release of newly synthesized IGFBP, the effect of a protein synthesis inhibitor was examined. When cycloheximide (1 ng/m\) was added to the cell cultures, there was no apparent IGFBP activity in 24-h conditioned medium in either the absence or presence of insulin (Table 2). The cells retained viability, since replacement of the medium with fresh SFM resulted in complete recovery of basal and insulin-stimulated IGFBP production. Therefore, bovine fibroblasts appear to actively synthesize and secrete IGFBPs, and this synthesis is regulated by insulin.
B
30-
21.5Serum
Molecular forms of bovine fibroblast IGFBP
To identify the particular IGFBP species secreted by bovine fibroblasts, Western ligand blotting was performed. Figure 1 shows the migration of IGFBPs from conditioned medium of cultured cells and from serum after sodium dodecyl sulfate-poly aery lamide gel electrophoresis (nonreducing conditions), transfer onto nitrocellulose, and incubation with [125I]IGF-I. In bovine and human serum, multiple bands were observed, two major bands of 42,000 and 38,000 Mr and additional bands at 34,000, 31,000, 28,000, and 24,000 Mr, similar to previously reported Western ligand blot patterns of mammalian serum and plasma (3, 4, 20). The 42,000 and 38,000 Mr bands have been shown to represent different glycoslation states of the binding subunit of the 150,000 Mr major plasma complex (3, 4). Conditioned medium from normal adult human fibroblasts contained the IGFBP forms seen in serum, as has been reported previously (18). Conditioned medium from SV40-transformed human fibroblasts also contained 42,000/38,000, 28,000, TABLE 2. Effect of cycloheximide on bovine fibroblast IGFBP production % of [125I]IGF-I bound" Treatment -INS +INS Recovery -INS +INS
Control
CX
15 30
5 7
10 29
11 27
3141
HF CM
I I SV40 HF CM
FIG. 1. Western ligand blot of IGFBPs. Serum and conditioned medium samples were separated on a 7.5-15% gradient sodium dodecyl sulfate-polyacrylamide gel under nonreducing conditions, electroblotted onto nitrocellulose filters, and after incubation with [126I]IGF-I, identified by autoradiography. +, Treatment with 100 nM insulin. BF CM, Bovine fibroblast 72-h conditioned medium; Serum B, Bovine; H, human; HF CM, human fibroblast 72-h conditioned medium; SV40 HF CM, SV40-transformed human fibroblast 48-h conditioned medium.
and 24,000 Mr IGFBPs, with the 24,000 Mr form being predominant. Bovine fibroblasts secreted low levels of 24,000, 28,000, and 34,000 Mr IGFBPs under control conditions (Fig. 1). In general, the 24,000 Mr species was the major IGFBP in the unstimulated state. Treatment of bovine fibroblast cultures with 100 nM insulin resulted in a relative 2- to
4-fold increase in these three IGFBP forms in the conditioned medium. Moreover, in the presence of insulin, bovine fibroblasts secreted significant amounts of 42,000 and 38,000 Mr IGFBP subunits, which were undetectable under basal conditions. In comparison, addition of 100 nM insulin to normal adult human fibroblasts or SV40transformed human fibroblasts resulted in little or no apparent change in the intensity or pattern of secreted IGFBPs. Regulation of IGFBP by insulin and IGF-I
Bovine fibroblasts were washed, preincubated in SFM for 24 h, and changed to SFM with or without insulin (1 tig/ml) in the presence or absence of cycloheximide (CX; 1 /ig/ml). After a 24-h treatment period, conditioned medium was collected and the cells were washed twice and incubated in SFM for a 6-h washout period. Medium was then replaced with SFM with or without insulin, and the conditioned medium was collected after a 24-h recovery period. 0 Results are expressed as a percentage of the total [126I]IGF-I added that is specifically bound. Values for 10 /xl conditioned medium samples are presented.
Figure 2 presents a time course for IGFBP production2 by bovine fibroblasts. Over 96 h in serum-free culture, bovine fibroblasts constitutively secreted 24,000, 28,000, and 34,000 Mr IGFBP forms. Treatment with insulin (100 nM) increased the production of all three forms at every time point, with a maximal effect seen by 72 h. Densitometic analysis of the band intensities of two separate experiments indicated that the 24,000, 28,000, and 34,000 Mr bands were increased 1.6- and 3.4-fold, 3.1- and 4.2-fold, and 1.8- and 24-fold, respectively, by 2
In this paper, production is used to denote a net effect, since potential degradation of IGFBPs was not determined.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
3142
REGULATION OF BOVINE FIBROBLAST IGFBP
Mr x10~ 3
Endo • 1990 Vol 126 • No 6
Mrx10-3
46-
46-
30-
3021.5-
21.5-
24h
48h
72h
96h
96h
INS (nM)
IGF-I (nM)
FIG. 2. Western ligand blot: time course of insulin (INS) and IGF-I stimulation. Conditioned medium samples from bovine fibroblasts incubated for the indicated times without or with 100 nM insulin or 10 nM IGF-I were analyzed by Western ligand blotting, as described in Fig. 1.
FIG. 3. Western ligand blot: effects of insulin (INS) and IGF-I. Conditioned medium from bovine fibroblast cultures untreated (C, control) or treated with various concentrations of insulin or IGF-I for 24 h were analyzed by Western ligand blotting, as described in Fig. 1.
72 h of insulin treatment. Moreover, insulin treatment for 24 h induced the apperance of a 42,000/38,000 Mr IGFBP form, which increased progressively over 96 h. There was no 42,000/38,000 Mr doublet evident in control culture medium, even with a 96-h incubation. IGF-I also increased the production of 24,000, 28,000, and 34,000 Mr IGFBPs and induced the secretion of 42,000/37,000 Mr IGFBP by bovine fibroblasts (Fig. 2B). The effects of IGF-I (10 nM) were equivalent to those of insulin (100 nM), except that an additional 31,000 Mr IGFBP band was sometimes observed with IGF-I treatment. Figure 2B presents a Western ligand blot of 96-h conditioned medium, but the effects of insulin and IGFI were also similar at 24, 48, and 72 h (data not shown). When [125I]IGF-II was employed as the radioligand, the only observed change was a relative 2-fold increase in the intensity of the 42,000/38,000 Mr bands from conditioned medium samples of insulin- and IGF-Itreated bovine fibroblasts and serum, indicating a slight preference for IGF-II by the 42,000/38,000 Mr IGFBP form, as has been reported previously for IGFBP-3 (21). The 24,000, 28,000, and 34,000 Mr IGFBP species appeared to bind [125I]IGF-I and [125I]IGF-II equivalent^ (data not shown). By Western ligand blot analysis, the dose-response effects of insulin and IGF-I on the regulation of bovine fibroblast IGFBP production were compared. Conditioned medium was collected 24 or 48 h after the addition of various concentrations of the two peptides, and the results of one of three experiments are presented in Fig. 3. Insulin and IGF-I, at low concentrations, were able to augment synthesis of the 24,000, 28,000, and 34,000 Mr forms of IGFBPs and induce secretion of the 42,000/ 38,000 Mr species. A stimulatory effect on all IGFBP forms was seen at 0.01 nM insulin. A maximal effect on
24,000, 28,000, and 34,000 Mr IGFBP production was observed with 1-10 nM insulin. IGF-I showed similar dose-response characteristics, except the magnitude of the response was less obvious at 0.01 and 0.1 nM IGF-I. A distinct stimulatory effect of 0.01 nM IGF-I was seen in one of the three experiments. In all three experiments, insulin and IGF-I at 1 nM had near-equivalent effects on IGFBP production by bovine fibroblasts. Furthermore, insulin and IGF-I treatment resulted in striking dosedependent increases in 42,000/38,000 Mr IGFBP secretion. Similar dose-response effects of insulin and IGF-I on IGFBP production were seen with a 48-h incubation period (data not shown). IGF-I and insulin appear to specifically affect IGFBP production independent of changes in general protein synthesis. Whereas IGF-I and insulin at 1 nM have similar near-maximal effects on IGFBP secretion, half-maximal effects on aminoisobutyric acid uptake in bovine fibroblasts were found with 2 and 20 nM IGF-I and insulin, respectively (data not shown). Quantitation of 42,000/38,000 Mr IGFBP production IGFBP induction by insulin and IGF-I in bovine fibroblast cultures was quantitated using purified bovine 42,000/38,000 Mr IGFBP (>98% pure) and Western ligand blotting. Conditioned medium samples from bovine fibroblasts treated with 10 nM insulin or IGF-I were run in parallel with known concentrations (5-160 ng/ml) of pure bovine 42,000/38,000 Mr IGFBP. Data were obtained by densitometric scanning of films from different exposure times, and linear dependence of optical density on the amount of IGFBP in the sample was validated. By densitometric analyses of band intensities and construction of standard curves for three separate experiments, it could be estimated that confluent cultures of
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP
3143
bovine fibroblasts (3.0 ± 0.3 x 105 cells) secrete 32 ± 2 and 36 ± 7 ng/ml 42,000/38,000 Mr IGFBP with 24 h of insulin or IGF-I stimulation, and 87 ± 14 and 124 ± 53 ng/ml after 96 h of insulin or IGF-I stimulation, respectively. Bovine serum contained 6.3 ± 0.6 /ug/ml and human serum contained 8.2 ± 0.8 /xg/ml 42,000/38,000 Mr IGFBP. Serum concentrations may be underestimated because of the labeling intensities of the bands.
matography fractions, 2) Western ligand blot analysis of neutral chromatography fractions, and 3) preincubation of conditioned medium with [125I]IGF-I before neutral chromatography. Parallel analyses of serum samples demonstrated 150,000 Mr IGFBP complexes under these conditions (data not shown).
Neutral chromatography of bovine fibroblast-conditioned medium
Bovine fibroblasts in culture synthesize and secrete low Mr IGFBPs which are precisely controlled by insulin and IGF-I in a time- and dose-dependent manner. Under basal serum-free conditions, bovine fibroblasts produce IGFBPs with Mr of 24,000, 28,000, and 34,000. The 24,000 Mr IGFBP was a predominant secretory product of bovine fibroblasts, as has been shown for normal adult human fibroblasts (18), transformed human fibroblasts, L6 rat muscle cells (17), and human breast cancer cells (22). This IGFBP form has not been sequenced as yet, but appears to be distinct from IGFBP1, IGFBP-2, and IGFBP-3 (18, 22, 23). The 34,000 Mr form is similar in size to the binding protein designated IGFBP-2, purified from the conditioned medium of MDBK and BRL3A cell cultures (8, 9, 24). This 34,000 Mr form secreted by cultured bovine fibroblasts did not exhibit a binding preference for IGF-II over IGF-I, which is characteristic of IGFBP-2 (8, 24), although relative affinities for the two peptides were not directly compared. The 28,000 Mr form is the same size as IGFBP-1 secreted by HepG2 cells. However, the secretion of bovine fibroblast 28,000 Mr IGFBP was stimulated by insulin, whereas insulin inhibits HepG2 IGFBP-1 production (18, 25). The 28,000 Mr form in human fibroblast-conditioned medium was also previously shown to be distinct from IGFBP-1 by immunological and physiological criteria (18). Whether these 34,000 and 28,000 Mr IGFBPs are additional IGFBP species or posttranslational modifications of one of the other forms remains to be determined. Production of the 34,000,28,000, and 24,000 Mr IGFBP forms by bovine fibroblasts was positively modulated by both insulin and IGF-I at subnanomolar concentrations. McCusker and Clemmons (17) similarly reported that insulin and IGF-I increased the secretion of 31,000 and 24,000 MJGFBPs in rat 1-6 muscle cells. In comparison, we have found the opposite effect of insulin in HepG2 cells, where insulin markedly and rapidly suppressed IGFBP-1 synthesis (18, 25). Likewise, IGFBP-1 concentrations in vivo are inversely correlated with circulating insulin concentrations (26). Clearly, insulin and IGF-I have differential effects on the various IGFBP species.
Conditioned medium samples from bovine fibroblasts stimulated by insulin (100 nM) and bGH (1 ng/m\), individually or in combination, were subjected to neutral chromatography on a Superose-12 column. Profiles of IGFBP activity recovered after fractionation are shown in Fig. 4. In medium from insulin-stimulated cells, a single peak of IGFBP activity eluted between albumin and ribonuclease-a markers (approximate Mr = 20,00050,000). Conditioned medium from bovine fibroblasts stimulated by insulin in the presence of bGH had an elution profile identical to that of insulin stimulation alone, with a single peak of IGFBP activity at 20,00050,000 Mr. There was no evidence of IGFBP activity in fractions corresponding to Mr greater than 60,000. In contrast, chromatographed human serum had two peaks of IGFBP activity, a high Mr peak eluting between /?amylase and albumin (approximate Mr = 150,000) and a 20,000-50,000 Mr peak. Conditioned medium from bovine fibroblasts treated for 72 h with IGF-I (10 nM) and bGH (1 Mg/ml) was also chromatographed under neutral conditions, as described above. No IGF-binding activity could be detected in high Mr fractions, as assessed by 1) assay of IGF-binding activity after acid chromatography of the neutral chro70 i-
60 T3
c o (D
50 40 30 20 10
0110
15
20
25
30
35
40
Fraction number FIG. 4. Superose-12 chromatography at neutral pH. Conditioned medium from bovine fibroblasts treated with 100 nM insulin (O) or insulin plus 1 fig/ml GH (A) and human serum samples (•) were gel filtered on a Superose-12 column in 50 mM Tris, 150 mM NaCl, and 0.5% BSA, pH 7.4. Fractions were assayed for IGFBP activity by charcoal separation ligand binding. The arrows indicate the elution volume of proteins used as Mr markers (xiO3). Vo, Void volume.
Discussion
Of particular interest, insulin and IGF-I specifically and markedly induced the secretion of the 42,000/38,000 Mr form of IGFBP, which corresponds in size to the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
3144
REGULATION OF BOVINE FIBROBLAST IGFBP
major IGFBP species in mammalian serum. This IGFBP is analogous to IGFBP-3, the acid-stable binding subunit of the 150,000 Mr complex in human plasma (10, 21). Confluent cultures of bovine fibroblasts secrete approximately 30 ng/24 h 42,000/38,000 Mr IGFBP after insulin or IGF-I treatment, as determined from a standard curve using known quantities of purified bovine 42,000/38,000 Mr IGFBP. With this method, bovine and human serum was estimated to contain approximately 6-8 ixg/m\ 42,000/38,000 Mr IGFBP, which agrees with measurements of this IGFBP form in human plasma at 6-9 ng/ ml using a specific RIA (27). Whether in vivo this form acts primarily at the site of synthesis to modulate IGF action or contributes to circulating IGFBP concentrations remains to be investigated. In human plasma, levels of the 42,000/38,000 Mr IGFBP species reflect GH status (27). The 42,000/38,000 Mr IGFBP secreted by cultured human fibroblasts was also found to be GH dependent (18). In contrast, GH had no effect on IGFBP production in the bovine fibroblast system. Instead, insulin and IGF-I potently and specifically induced the synthesis of 42,000/38,000 Mr IGFBP by bovine fibroblasts. Species differences may account for differential regulatory mechanisms for 42,000/38,000 Mr IGFBP in the human and bovine systems. However, bovine IGF-I has been sequenced and found to be identical to human IGF-I (28). Therefore, bovine and human IGFBP forms and their control mechanisms would be expected to be highly homologous, if not identical. Alternatively, IGF-I may be the principal regulator of the 42,000/38,000 Mr class of IGFBPs, and GH could have indirect effects. In plasma, IGF-I concentrations, which are GH dependent (2), could be responsible for the observed GH dependency of 42,000/38,000 Mr IGFBP. Similarily, human fibroblasts have been shown to secrete IGF-like peptides, the synthesis of which can be stimulated with GH treatment (29-31). This could account for 42,000/38,000 Mr IGFBP production by human fibroblasts in the basal state and for GHdependent increases in IGFBP synthesis (18). It is of note that cultured bovine fibroblasts do not appear to secrete IGF peptide (my unpublished observations). Consistent with this hypothesis, Zapf et at. (32) found that IGF-I induced 42,000-49,000 Mr IGFBP in plasma of hyppophysectomized and diabetic rats. Furthermore, the ontological appearance of 43,000/39,000 Mr IGFBP subunits corresponded in time to the appearance of IGF-I in fetal porcine serum (20), and a correlation between the 42,000-49,000 Mr cluster and IGF-I status was observed in rat osteoblastic cells in vitro (15). In addition to IGF-I, insulin appears to play an important role in regulating IGFBP production, as seen in the bovine fibroblast model system, and, thus, may coordinate IGF bioactivity with nutritional status.
Endo • 1990 Vol 126 • No 6
Although GH may not be the primary regulator of 42,000/38,000 Mr IGFBP synthesis, the findings of Zapf et al. (32) indicated that GH was necessary for association of these binding subunits into the 150,000 Mr circulating IGFBP complex. This theoretically could come about by GH-regulated production of the acid-labile subunit, which, along with IGF peptide, is necessary for complex formation (33). Although bovine fibroblasts secrete large amounts of 42,000/38,000 Mr IGF-binding subunits in response to insulin and IGF-I, in the presence of GH these binding subunits did not form any large Mr IGFBP complex, as could be found in serum. Therefore, bovine fibroblasts in culture may not be responsive to GH or may be unable to produce the acid-labile subunit necessary for complete association. Thus, physiological concentrations of insulin and IGFI stimulate IGFBP production in cultured bovine fibroblasts. The mechanisms by which the different IGFBP forms affect IGF endocrine and autocrine/paracrine functions are unknown; however, both inhibiting and potentiating effects have been noted in vitro (11-14). The distribution of each IGFBP form and the hormonal and nutritional control of their synthesis and secretion may be important determinants of IGF action in normal and abnormal physiology.
Acknowledgments I would like to thank Laurie Bale and Jay Clarkson for their excellent technical assistance.
References 1. Froesch ER, Schmid C, Schwander J, Zapf JA 1985 Actions of insulin-like growth factors. Rev Physiol 47:443 2. Daughaday WH, Rotwein P 1989 Insulin-like growth factors I and II. Peptide, messenger ribonucleic acid and gene structures, serum, and tissue concentrations. Endocr Rev 10:68 3. Hardouin S, Hossenlopp P, Segovia B, Seurin D, Portolan G, Lassare C, Binoux M 1987 Heterogeneity of insulin-like growth factor binding proteins and relationships between structure and affinity. I. Circulating forms in man. Eur J Biochem 170:121 4. Yang YW-H, Wang J-F, Orlowski CC, Nissley SP, Rechler MM 1989 Structure, specificity, and regulation of the insulin-like growth factor-binding proteins in adult rat serum. Endocrinology 125:1540 5. Povoa G, Enberg G, Jornvall H, Hall K 1984 Isolation and characterization of a somatomedin binding protein from mid-term amniotic fluid. Eur J Biochem 144:199 6. Hossenlopp P, Seurin D, Segovia-Quinson B, Binoux M 1986 Identification of an insulin-like growth factor binding protein in human cerebrospinal fluid with a selective affinity for IGF-II. FEBS Lett 208:439 7. Lee Y-L, Hintz RL, James PM, Lee PDK, Shively JE, Powell DR 1988 Insulin-like growth factor (IGF) binding protein complementary deoxyribonucleic acid from human HepG2 hepatoma cells: predicted protein sequence suggests an IGF binding domain different from those of the IGF-I and IGF-II receptors. Mol Endocrinol 2:404 8. Binkert C, Landwehr J, Mary J-L, Schwander J, Heinrich G 1989 Cloning, sequence analysis and expression of a cDNA encoding a novel insulin-like growth factor binding protein (IGFBP-2). EMB0
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
REGULATION OF BOVINE FIBROBLAST IGFBP J 8:2497 9. Brown AL, Chiariotti L, Orlowski CC, Mehlman T, Burgess WH, Ackerman EJ, Bruni CB, Rechler MM 1989 Nucleotide sequence and expression of a cDNA clone encoding a fetal rat binding protein for insulin-like growth factors. J Biol Chem 264:5148 10. Wood WI, Cachianes G, Henzel WJ, Winslow GA, Spencer SA, Hellmiss R, Martin J, Baxter RC 1988 Cloning and expression of the growth hormone-dependent insulin-like growth factor-binding protein. Mol Endocrinol 2:1176 11. Ritvos 0, Ranta T, Jalkanen J, Suikkari A-M, Voutilainen R, Bohn H, Rutanen E-M 1988 Insulin-like growth factor (IGF) binding protein from human decidua inhibits the binding and biological action of IGF-I in cultured choriocarcinoma cells. Endocrinology 122:2150 12. Busby WH, Klapper DG, Clemmons DR 1988 Purification of a 31,000-dalton insulin-like growth factor binding protein from human amniotic fluid: isolation of two forms with different biologic actions. J Biol Chem 263:14203 13. Ross M, Francis GL, Szabo L, Wallace JC, Ballard JF 1989 Insulinlike growth factor (IGF) binding proteins inhibit the biological activities of IGF-1 and IGF-2 but not des-(l-3)-IGF-l. Biochem J 258:267 14. DeMellow JSM, Baxter RC 1988 Growth hormone-dependent insulin-like growth factor (IGF) binding protein both inhibits and potentiates IGF-I-stimulated DNA synthesis in human skin fibroblasts. Biochem Biophys Res Commun 156:199 15. Schmid C, Zapf J, Froesch ER 1989 Production of carrier proteins for insulin-like growth factors (IGFs) by rat osteoblastic cells. FEBS Lett 244:328 16. Moses AC, Freinkel AJ, Knowles BB, Aden DP 1983 Demonstration that a human hepatoma cell line produces a specific insulinlike growth factor carrier protein. J Clin Endocrinol Metab 56:1003 17. McCusker RH, Clemmons DR 1988 Insulin-like growth factor binding protein secretion by muscle cells: effect of cellular differention and proliferation. J Cell Physiol 137:505 18. Conover CA, Liu F, Powell D, Rosenfeld RG, Hintz RL 1989 Insulin-like growth factor binding proteins from cultured human fibroblasts: characterization and hormonal regulation. J Clin Invest 83:852 19. Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M 1986 Analysis of serum insulin-like growth factor binding proteins using Western blotting: use of the method for titration of the binding proteins and competitive binding studies. Anal Biochem 154:138 20. McCusker RH, Campion DR, Jones WK, Clemmons DR 1989 The
21. 22. 23.
24.
25. 26.
27. 28. 29. 30. 31. 32.
33.
3145
insulin-like growth factor binding proteins of porcine serum: endocrine and nutritional regulation. Endocrinology 125:501 Martin JL, Baxter RC 1986 Insulin-like growth factor-binding protein from human plasma: purification and characterization. J Biol Chem 261:8754 DeLeon DD, Wilson DM, Bakker B, Lamson G, Hintz RL, Rosenfeld RG 1989 Characterization of insulin-like growth factor binding proteins from human breast cancer cells. Mol Endocrinol 3:567 Rosenfeld RG, Pham H, Oh Y, Lamson G, Guidice LC 1990 Identification of insulin-like growth factor binding protein-2 (IGFBP-2) and a low molecular weight IGFBP in human seminal plasma. J Clin Endocrinol Metab 70:551 Szabo L, Mottershead DG, Ballard JF, Wallace JC 1988 The bovine insulin-like growth factor (IGF) binding protein purified from conditioned medium requires the N-terminal tripeptide in IGF-I for binding. Biochem Biophys Res Commun 151:207 Conover CA, Lee PDK 1990 Insulin regulation of IGF binding protein production in cultured HepG2 cells. J Clin Endocrinol Metab 70:1062 Suikkari AM, Koivisto VA, Rutanen EM, Yki-Jarvinen H, Karonen SL, Seppala M 1988 Insulin regulates the serum levels of low molecular weight insulin-like growth factor-binding protein. J Clin Endocrinol Metab 66:266 Baxter RC, Martin JL 1986 Radioimmunoassay of growth hormone-dependent insulinlike growth factor binding protein in human plasma. J Clin Invest 78:1504 Honegger A, Humbel RE 1986 Insulin-like growth factors I and II in fetal and adult bovine serum. J Biol Chem 261:569 Atkison PR, Weidman ER, Blaimick B, Bala RM 1980 Release of somatomedin-like activity by cultured WI-38 human fibroblasts. Endocrinology 106:2006 Clemmons DR, Underwood LE, Van Wyk JJ 1981 Hormonal control of immunoreactive somatomedin production by cultured human fibroblasts. J Clin Invest 67:10 Conover CA, Baker BK, Hintz RL 1989 Cultured human fibroblasts secrete insulin-like growth factor IA prohormone. J Clin Endocrinol Metab 69:25 Zapf J, Hauri C, Waldvogel M, Futo E, Hasler H, Binz K, Guler HP, Schmid C, Froesch ER 1989 Recombinant human insulin-like growth factor I induces its own specific carrier protein in hypophysectomized and diabetic rats. Proc Natl Acad Sci USA 86:3813 Baxter RC, Martin JL, Beniac VA 1989 High molecular weight insulin-like growth factor binding protein complex: purification and properties of the acid-labile subunit from human serum. J Biol Chem 264:11843
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 07:18 For personal use only. No other uses without permission. . All rights reserved.
