Autocrine role of IGF-II in proliferation of human adrenocortical carcinoma NCI H295R cell line
A Logié, N Boulle, V Gaston, L Perin, P Boudou1, Y Le Bouc and C Gicquel
Laboratoire d’Explorations Fonctionnelles Endocriniennes, Hôpital Trousseau AP-HP, Paris, France 1 Laboratoire de Biologie Hormonale, Hôpital Saint-Louis, AP-HP, Paris, France
(Requests for offprints should be addressed to C Gicquel, Laboratoire d’Explorations Fonctionnelles Endocriniennes, Hôpital Trousseau, 26 Avenue Arnold Netter, 75012 Paris, France)
ABSTRACT
In adrenocortical tumors, the malignant phenotype is associated with rearrangements (paternal iso- disomy) at the 11p15 locus and IGF-II gene overexpression, strongly suggesting that the IGF system is a major determinant of adrenocortical tumor progression.
The aim of this study was to validate an in vitro model for investigating the involvement of the IGF system in adrenocortical tumorigenesis. We analyzed the production of IGF mRNA and proteins, IGF-binding proteins (IGFBPs) and IGF receptors by the NCI H295R cell line, which is derived from a human adult adrenocortical carcinoma.
H295R cells were shown to proliferate for a long period (26 days) in the absence of serum or any added growth factor. Northern blot analyses showed high IGF-II mRNA contents in H295R cells. The cells secreted large amounts of IGF-II protein (14 ng/106 cells per 48 h) although no IGF-I protein was detected.
Western ligand blot analyses of conditioned media detected the presence of large amounts of a 34 kDa protein, which was identified as IGFBP-2 by immunoblotting.
The presence of high-affinity binding sites for IGF-I and IGF-II on H295R cells was shown by binding experiments using radiolabeled IGFs and confirmed by reverse transcription PCR analyses showing type 1 and type 2 IGF receptors. Proliferation of H295R cells was inhibited by anti-IGF-II antibody (45%) and by anti-type 1 IGF receptor antibody (53%) indicating that IGF-II is an autocrine growth factor for these cells and that its effects are, at least in part, mediated by the type 1 IGF receptor.
These findings confirm the involvement of the IGF system in adrenocortical tumors and suggest that the H295R cell line is a suitable in vitro model for studying the molecular mechanisms of adrenocortical tumor proliferation.
Journal of Molecular Endocrinology (1999) 23, 23-32
INTRODUCTION
Malignant adrenocortical carcinomas are rare tumors with a poor prognosis. The molecular mechanisms involved in adrenocortical tumori- genesis are still poorly understood. Recent studies have focused on alterations of the insulin-like growth factor (IGF) system associated with these tumors. For instance, there are abnormalities at the 11p15 region, where the IGF-II gene maps, in more than 90% of malignant adrenocortical tumors but in less than 10% of benign tumors (Gicquel et al. 1994, 1997). The tumors with these abnormalities exhibit strong overexpression of the IGF-II gene (Ilvesmäki et al. 1993, Gicquel et al. 1994, 1997)
and large amounts of IGF-II protein (Boulle et al. 1998). The type 1 IGF receptor (Weber et al. 1997) and the IGF-binding protein-2 (IGFBP-2) (Boulle et al. 1998) are also specifically overproduced in malignant adrenocortical tumors. These findings strongly implicate the IGF system in adrenocortical tumor progression.
The IGF system comprises several elements. The IGFs, IGF-I and IGF-II, are small polypeptides produced in various tissues and cell cultures. They have endocrine and auto/paracrine modes of action (Jones & Clemmons 1995). Two structurally different IGF receptors have been described, the type 1 IGF receptor mediating most effects of the IGFs and the IGF-II/mannose-6-phosphate
Online version via http://www.endocrinology.org
(IGF-II M6P) receptor the function of which should be the internalization and subsequent degradation of IGF-II (Jones & Clemmons 1995). IGF-I and IGF-II can also bind with high affinity to IGFBPs. Six high-affinity IGFBPs have been described to date (Jones & Clemmons 1995). These IGFBPs modulate the effects of IGFs either positively or negatively depending on their abundance, their affinity for the growth factors and their cellular localization.
IGF-II has been implicated in the growth of various tumors including Wilms’ tumors, hepato- mas, colon carcinomas and pheochromocytomas, suggesting that IGF-II plays a central role in tumorigenesis (Christofori et al. 1994, 1995, Werner & LeRoith 1996). Similarly, IGF-II may also be involved in adrenocortical tumors. However, direct proof for this role is lacking. We therefore tested an in vitro adrenocortical tumor model to investigate the autocrine role of IGF-II in the growth of such tumors. We took advantage of a cell line (NCI H295R) established from a human malignant adrenocortical tumor (Gazdar et al. 1990, Rainey et al. 1994).
In this paper, we show that the H295R cell line is a suitable in vitro model for studying the role of the IGF system in adrenocortical tumorigenesis and demonstrate that IGF-II is an auto/paracrine factor involved in the proliferation of adrenocortical tumor cells.
MATERIALS AND METHODS
Cell cultures
NCI H295R cells (National Cancer Institute, Bethesda, MD, USA) were grown in a 1:1 mixture of Dulbecco’s modified eagle’s medium (DMEM) and Ham’s F12 medium (Sigma Chemical Company, St Louis, MO, USA) supplemented with transferrin (5 µg/ml; Sigma), sodium selenite (5 ng/ml; Sigma), L-glutamine (2.5 mM; Gibco BRL, Paisley, Strathclyde, UK) and antibiotics (50 µg/ml strepto- mycin, 50 IU/ml penicillin; Gibco BRL) with or without 2% Ultroser G (Biosepra, Marlborough, MA, USA) at 37 °℃ in a 5% CO2 atmosphere. After reaching confluence, cells were split using 0.05% trypsin with 0-02% EDTA, and they were used at passages three to six for experiments.
For proliferative analyses, cells were grown for up to 26 days in 78 cm2 dishes (4 x 103 cells/cm2) (Nunc, Roskilde, Denmark) in serum-free medium, which was changed every 48 h. The medium was collected after various times of culture for IGF and IGFBP assays and the cells were counted or treated for RNA extraction.
For antibody experiments, cells were grown in serum-free medium in 96-well dishes (105 cells/cm2) (Nunc). After 4 days of culture, the medium was renewed (time 0 of the experiment) and replaced every day with or without the indicated concen- trations of monoclonal anti-type 1 IGF receptor antibody (1, 2, 5, 10 and 15 µg/ml) (aIR3; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or anti-IGF-II antibody (1, 2, 5, 10 and 30 µg/ml) (UBI, Lake Placid, NY, USA). Non-specific antibody (mouse IgG1k (MOPC-21); Sigma) was used as a control at the same concentrations as either aIR3 or anti-IGF-II antibodies. Cells fed only with serum-free medium were also used as controls. Ninety-six hours after the addition of antibodies, the incorporation of [3H]thymidine was assayed as described below.
[3H ] Thymidine incorporation [3H]Thymidine incor- poration was evaluated after a 48 h incubation period with 18.5 x 104 Bq/ml [3H]thymidine (specific activity 3.1 × 1012 Bq/mmol) (Amersham International, Amersham, Bucks, UK). Cells were washed three times with PBS and lysed in 100 ul NaOH 0.6 M at 37 ℃. The suspension (50 ul) was counted in a scintillation counter (Liquid Scintil- lation Counter 1209 Rackbeta; Pharmacia, Uppsala, Sweden).
IGF assays
The methods used have been described in detail elsewhere (Hardouin et al. 1989). Lyophilized samples corresponding to 2.5-12.5 ml culture medium were gel filtered in 1 M acetic acid on columns of Ultrogel AcA 54 (Biosepra) to separate IGFs from IGFBPs. The eluates were lyophilized and, before being assayed, desalted on Sephadex G-25 disposable columns (Pharmacia, Sweden) in assay buffer. IGF-I was assayed by RIA using an anti-IGF-I antibody (gift from Dr J Closset, Liege, Belgium). IGF-II was measured by competitive protein binding assay using IGFBPs extracted from cerebrospinal fluid, which have a selective affinity for IGF-II (Binoux et al. 1986).
The IGF preparations used for radiolabeling and as standards were recombinant human IGF-I and IGF-II, kindly provided by Ciba Geigy (Basel, Switzerland).
Steroid assays
Steroid production by H295R cells was measured in conditioned media from cells grown for 48 h in serum-free medium. Progesterone, dehydro- epiandrosterone sulfate (DHEA-S), cortisol and
Journal of Molecular Endocrinology (1999) 23, 23-32
aldosterone levels were determined using the 125 I-progesterone Coatria kit (Biomerieux, Lyon, France), the 125I-DHEA-S kit (Immunotech, Marseille, France), the Gamma Coat 125I-cortisol kit (Incstar Corp., Stillwater, MN, USA) and the Coat-A-Count 125I-aldosterone kit (Diag- nostic Products Corp., Los Angeles, CA, USA) respectively. Moreover, pregnenolone (PREG), 17-hydroxypregnenolone (17OH-PREG), 17- hydroxyprogesterone (17OH-P), dehydroepiandro- sterone (DHEA), delta-4 androstenedione, 11-deoxycorticosterone (DOC), 11-deoxycortisol (11-DOF), corticosterone, 11ß-OH delta-4 andro- stenedione (11ßOH-A) and testosterone levels were quantified by specific RIA methods involving chromatographic purification of medium extracts as previously described (Fiet et al. 1994).
Western ligand and immunoblotting
Western ligand and immunoblotting were performed as previously described (Hossenlopp et al. 1986, Lalou & Binoux 1993). Briefly, conditioned media were concentrated by lyo- philization, denatured at 100 ℃ for 2 min and subjected to 11% PAGE under non-reducing conditions, and then electrotransferred onto nitro- cellulose (BA 85; Schleicher and Schuell, Dassel, Germany). The different IGFBP species were detected by: (i) incubation with a mixture of 125I-IGF-I and 125I-IGF-II (5 x 105 c.p.m. each) at 4 ℃ for 48 h, followed by autoradiography (ligand blotting); and (ii) incubation with specific antibodies: anti-human IGFBP-1 serum and anti-human IGFBP-3 serum kindly provided by Dr Binoux (Inserm U 142, Paris, France), anti-bovine IGFBP-2, anti-human IGFBP-4 and anti-human IGFBP-5 polyclonal antibodies pur- chased from Upstate Biotechnology (Lake Placid, NY, USA). Anti-human IGFBP-6 polyclonal antibody was obtained from A F Schützdeller (Tübingen, Germany). After incubation with a second anti-IgG antibody coupled to horseradish peroxidase (BIO Source, Camarillo, CA, USA), the complexes were visualized by chemiluminescence using the enhanced chemiluminescence Western blotting detection system (Amersham).
For IGF-II immunoblotting, 800 ul conditioned medium (cells cultured for 48 h in serum-free medium) were lyophilized, loaded on a 12.5% non-denaturing gel and transferred to nitro- cellulose. The blot was probed with an anti-rat IGF-II monoclonal antibody (1/500) purchased from Upstate Biotechnology. This antibody is specific for rat and human IGF-II and shows less than 10% cross-reactivity with human IGF-I.
Ligand binding assays for IGF cell surface receptors
Cells were plated out in 24-well dishes (5 x 105 cells/cm2) (Nunc) and cultured for 48 h in medium with 2% Ultroser G. The cells were rinsed three times and fed with 1 ml/well fresh serum-free medium. After 24 h, the cell monolayers were placed on ice and washed with 1 ml binding medium (DMEM/F12 (1:1), 25 mM Hepes pH 7-4, 2.5 mM L-glutamine, 1 mg/ml BSA (Pasteur Biomérieux, Lyon, France)), preincubated for 30 min with unlabeled ligand and then incubated in a total volume of 0.25 ml binding medium for 4 h at 4 ℃ with 125I-IGF-I (2.8x 104 c.p.m.) or 125I-IGF-II (2.3x 104 c.p.m.) (7 x 105 c.p.m./pmol IGF-I; 5.3 x 105 c.p.m./pmol IGF-II), with vari- ous concentrations of either IGF-I (0.01 nM to 10 µM), IGF-II (0.01 to 1000 nM), des(1-6)IGF-II (GroPep, Adelaide, Australia) (0-01-10 uM) or insulin (Sigma) (0.1 nM to 100 µM). At the end of incubation, cells were washed four times with Hanks’ solution (Sanofi Pasteur, Marnes la coquette, France) at 4 ℃ and then lysed at room temperature in 0.5 M NaOH. Cell-associated radioactivity was counted in a gamma counter.
Isolation of RNA and mRNA analysis
Total RNA was isolated as previously described (Chomczynski & Sacchi 1987). Briefly, cells were rinsed three times in PBS, lysed in 4 M guanidinium thiocyanate, 25 mM sodium citrate pH 7, 0.5% sarkosyl and 0.1 M ß-mercaptoethanol and RNA was extracted by the phenol-chloroform method. Seven micrograms of total RNA were loaded onto a 1·1% agarose-formaldehyde gel, then transferred to a Genescreen plus membrane (NEN Research Products, DuPont de Nemours & Co., Germany), and covalently bound to the membrane by UV irradiation. Northern blots were hybridized with cDNA probes for human IGF-II (Le Bouc et al. 1987), IGFBP-2 or IGFBP-3 (kindly provided by Dr Binoux, Inserm U 142, Paris, France).
Type 1 and type 2 IGF receptors expression was analyzed by reverse transcription PCR (RT-PCR). RNA samples (1 µg) were treated with DNase to eliminate any DNA contamination, reverse- transcribed using Moloney murine leukemia virus reverse transcriptase (Gibco) and resulting cDNA amplified by PCR. Sense and antisense primers for the type 1 IGF receptor were respectively 5’ AAC CAC GAG GCT GAG AAG CT and 5’ CAG CAT AAT CAC CAA CCC TC (Ilvesmäki et al. 1993). Oligonucleotide primers to type 2 IGF receptor were designed within the 3’ UTR of the human IGF-II
4
T
1
Cell number (x 106)
3
2
1
0
2
4
7 11 15 18 20 24 26 days
receptor (Hol et al. 1992). Sense and antisense pri- mers were respectively 5’ TTG CCG GCT GGT GAA TTC AA and 5’ GTA TCA TGA GAA CCT GAA GAG. DNase-treated RNA PCR products were used as control for DNA contamination of the RNA samples. Products of amplification were analyzed by electrophoresis on a 1.5% agarose gel.
RESULTS
H295R cells grow spontaneously in serum-free medium
Gazdar et al. (1990) established the H295 cell line from an invasive primary adrenocortical carcinoma. This cell line was the first to maintain the ability to produce all the adrenocortical steroids. Later, Rainey et al. (1994) selected a population of H295 cells with anchorage-dependent growth which they named the NCI H295R cell line. These adherent cells were used for the present study.
Preliminary studies with H295 cells showed that they had a long population doubling time (>96 h) and could only proliferate in medium supplemented with selenium, insulin and transferrin (Gazdar et al. 1990, Rainey et al. 1994). Because we wanted to study the role of the IGF system in H295R cell proliferation, it was important to verify the spontaneous growth of this cell line in the absence of any added growth factors, particularly insulin, which could interfere with the type 1 IGF receptor.
H295R cells were capable of spontaneous although slow growth in the absence of serum or added growth factors (Fig. 1). In the exponential phase, the number of cells doubled every 2.5 days after a 15 day latency period.
We also extensively studied the secretion of steroids by H295R cells growing in serum-free medium (Fig. 2). H295R cells were pluripotent cells and produced glucocorticoids, mineralocorticoids and androgens. The production of androgens was predominant, involving mainly the delta-4 pathway (i.e. progesterone, 17OH-P, androstenedione) when the delta-5 pathway was minor (i.e. PREG, 17OH-PREG, DHEA). High levels of DOC and 11-DOF and lower levels of corticosterone, cortisol and 11BOH-A suggested a potential alteration in the 11ß-hydroxylase activity.
IGF-II is produced by H295R cells
We previously showed high IGF-II mRNA and protein contents in malignant adrenocortical tumors (Gicquel et al. 1994, 1997, Boulle et al. 1998). Such IGF-II production would be expected from the H295R cell line as it was derived from a malignant human adrenocortical tumor. IGF-I and IGF-II proteins were measured in conditioned media from these cells cultured in serum-free medium (Fig. 3A). H295R cells secreted large amounts of IGF-II protein and the amount of growth factor produced per cell increased with time in culture (from 14 ng/10 cells per 48 h at day 2 to 72 ng/106 cells per 48 h at day 11). By contrast, IGF-I protein was not detected in conditioned media. The IGF-II assay used here allowed measurement of mature 7.5 kDa IGF-II (Binoux et al. 1986). To determine whether other forms of the growth factor were produced by H295R cells, aliquots of conditioned media were immunoblotted with a specific IGF-II antibody (Fig. 3B). Both the mature 7.5 kDa peptide and significant amounts of high molecular weight forms of IGF-II (from 8.5 to 24 kDa), presumably precursor forms, were detected.
Northern blotting analysis showed that the sizes of IGF-II mRNA (2.2, 4.8 and 6.0 kb) were the same as in normal adrenal tissues and tumors (Fig. 3C) (Gicquel et al. 1994). Kinetic analysis of IGF-II gene expression showed an increase in IGF-II mRNA during proliferation in serum-free medium (Fig. 3C).
IGFBPs secretion by H295R cells
IGFBPs are important regulators of IGF effects. We therefore examined the production of IGFBPs by H295R cells. Western ligand blotting detected a major band of 34 kDa in the culture medium of H295R cells that co-migrated with the 34 kDa band of a control human serum (Fig. 4A). After prolonged cell proliferation (18-20 days), a minor 39-42 kDa doublet was also detected (Fig. 4A).
CHOLESTEROL
P450 scc
PREG 9.2
DHEA 5.1
DHEA-S 324
P450 c17
17 OH-PREG 52.1
P450 c17 17-20 lyase
3BHSD
P 48.6
P450 c17
17 OH-P 88.6
P450 c17 17-20 lyase
P450 c21
DOC 66.7
11 DOF 1196
A
283
17 BHSD
T
13.4
P450 c11 B1
B
F
11 BOH-A 31.3
207
356
P450 c11 B2
ALDO 1.4
Immunoblotting, using specific antisera, ident- ified the 34 kDa band as IGFBP-2 (Fig. 4A) and the 39-42 kDa doublet as IGFBP-3 (Fig. 4B). In both cases, the IGFBP appeared as an intact protein by comparison with a control human serum and no proteolytic fragments were detected either with anti-IGFBP-2 or anti-IGFBP-3 antibodies. Specific antisera to IGFBP-1, -4, -5 and -6 failed to identify these IGFBPs in H295R-conditioned medium (data not shown). In agreement with the ligand blot data, Northern blot analyses showed that H295R cells contained mRNAs for IGFBP-2 (1·6 kb) and IGFBP-3 (2·6 kb) (Fig. 4C).
Binding of 125I-IGF-I and 125I-IGF-II to H295R cells
The next series of experiments investigated the existence of IGF binding sites on H295R cell surface. IGF-I binding is shown in Fig. 5A. The mean specific binding of 125I-IGF-I to H295R cells was 12.3 ± 0.3% (S.E.M.) and it was effectively displaced by unlabeled IGF-I with a 50% displace- ment (IC50) at 0.53 nmol/l. IGF-II and insulin competitively inhibited the binding of 125I-IGF-I in a decreasing order according to their affinity to the type 1 IGF receptor (5-fold and 1500-fold lower affinities for respectively IGF-II and insulin).
IGF-II binding is shown in Fig. 5B. The mean specific binding of 125I-IGF-II to H295R cells was 12.8 ± 0.4% and it was effectively displaced by unlabeled IGF-II with an IC50 of 1.7 nmol/l. Partial displacement of 125I-IGF-II was possible with IGF-I and insulin, confirming the presence of the type 1 IGF receptor. However, this displace- ment of 125I-IGF-II from the H295R cell surface by IGF-I and insulin was not complete, suggesting that these cells also express a specific IGF-II binding site different from the type 1 IGF receptor. The complete displacement of 125I-IGF-II by des(1-6)IGF-II excludes an IGFBP as the second IGF-II binding site and rather suggests that this site is the type 2 IGF receptor.
The presence of type 1 and type 2 IGF receptors on H295R cells was confirmed by RT-PCR analyses showing that both receptors are expressed by this cell line (Fig. 6).
IGF-II is involved in proliferation of H295R cells through the type 1 IGF receptor
The high levels of IGF-II expression and the presence of IGF-II binding sites on H295R cells are strong evidence for an autocrine role of IGF-II in H295R cell growth. To confirm this, [3H]thymidine incorporation as a measure of H295R cell prolifer- ation was studied in the presence of an anti-IGF-II
A
B
rhIGF-II
H295R cell
medium
malignant
AC tumor
hypoglycemic
tumor
IGF-II (ng/106 cells/48 h)
70
kDa
50
24
30
17
10
7.5
2
4
7
11
18 days
C
day
4
11
15
18
20
IGF- II
6 kb -
4.8 kb -
2.2 kb -
28S
antibody that specifically neutralizes IGF-II, and in the presence of the aIR3 antibody which blocks the type 1 IGF receptor. Dose-response curves for the effects of anti-IGF-II and aIR3 antibodies on [3H]thymidine incorporation are shown in Fig. 7.
Both the anti-IGF-II antibody and the aIR3 antibody inhibited H295R cell proliferation in a dose-dependent manner with maximal inhibitions of 45 and 53.4% respectively, compared with control cells (Fig. 7). This effect was specific and normal mouse IgG (1-30 µg/ml) had no effect on cell proliferation when compared with cells fed only with serum-free medium (data not shown). These results are direct evidence that IGF-II, acting, at
least in part, through the type 1 receptor, is involved in the autocrine proliferation of H295R cells.
DISCUSSION
There is increasing evidence that IGF-II plays a determinant role in tumor proliferation (reviewed in Werner & LeRoith 1996). This is particularly true for adrenocortical tumors. Study of the IGF system in these tumors indeed showed that strong over- expression of the IGF-II gene is a frequent feature of the malignant state (Ilvesmäki et al. 1993, Gicquel et al. 1994, 1997, Liu et al. 1995). The IGF-II gene
A
normal serum
4
11
15
18
20
day
kDa
42
39
34
WLB
WIB
B
normal serum
C
day
4
11
15
18
20
kDa
day 20
IGFBP-2
42-
1.6 kb -
39 -
IGFBP-3
30-
WIB
2.6 kb -
28S
maps to the 11p15 imprinted region and is expressed from the paternal allele (De Chiara et al. 1991, Ogawa et al. 1993, Rainier et al. 1993). The mechan- ism of IGF-II gene overexpression in malignant tumors involves pathological imprinting. Indeed, most tumors with overexpression of the IGF-II gene also exhibit paternal isodisomy (loss of the maternal- derived allele and duplication of the IGF-II active paternal allele) or, less frequently, loss of imprinting (with maintenance of the maternal allele but a paternal-like IGF-II gene expression pattern) (Gicquel et al. 1997). IGF-II mRNA is efficiently translated and malignant tumors contain large amounts of IGF-II protein, partly as precursor
forms (Boulle et al. 1998). IGF-I and IGF-II recep- tors are present in adrenal tissues and overexpression of the type 1 IGF receptor has been shown in adrenocortical carcinomas (Kamio et al. 1991, Ilvesmäki et al. 1993, Backlin et al. 1995, Weber et al. 1997). These previous studies all used frozen tumor tissues and give strong evidence for a role for IGF-II in adrenocortical tumor proliferation. However, no direct proof of the contribution of IGF-II is avail- able. To investigate the function of IGF-II in growth regulation of adrenocortical tumors, we used the H295R cell line (Rainey et al. 1994) derived from the H295 cell line established from an invasive hormonally active primary adrenocortical carcinoma
A
125|-IGF-1(% of maximal binding)
100
±
80
60
40
20
1
E
O
0
-11
-10
-9
-8
-7
-6
-5
-4
unlabeled ligand (Log mol/L)
B
1251-IGF-II(% of maximal binding)
100
80
60
車
40
C
1
20
Đ
0
-11
-10
-9
-8
-7
-6
-5
-4
unlabeled ligand (Log mol/L)
(Gazdar et al. 1990). Analysis of the IGF system of this cell line has not been described other than some preliminary results on IGF-II mRNA production (Staels et al. 1993).
H295R cells spontaneously grow in serum-free medium suggesting that they may secrete autocrine growth factors. Here we show that the IGF-II (and not the IGF-I) gene is strongly expressed in H295R cells as in malignant human adrenocortical tumors (Ilvesmäki et al. 1993, Gicquel et al. 1994). We also show that IGF-II mRNAs are efficiently translated, with accumulation of IGF-II protein in culture
media during cell proliferation, partly as precursor forms, as we previously showed in frozen adreno- cortical carcinomas (Boulle et al. 1998). During proliferation, both the content of IGF-II mRNA and the amount of secreted IGF-II protein increased, suggesting that IGF-II overexpression is determined at the transcriptional level. Analysis of IGFBP expression in H295R cells by Western ligand and immunoblotting detected IGFBP-2 and almost no other IGFBP. These results support previous data from frozen human adrenocortical tumors showing that tumors overexpressing the IGF-II gene (malignant tumors) have a high IGFBP-2 content (Boulle et al. 1998).
IGF binding assays indicate that IGF-II binds to two different IGF binding sites on the H295R cell surface. One of these IGF-II binding sites is probably the type 1 IGF receptor as shown by IGF-I binding experiments and RT-PCR analysis. The type 1 IGF receptor is generally responsible for the transmission of the growth-promoting activities of both IGF-I and IGF-II (Jones & Clemmons 1995). Moreover H295R cell proliferation is inhibited by anti-IGF-II and by aIR3 antibodies demonstrating that IGF-II stimulates cell growth by an autocrine mechanism through the type 1 IGF receptor. Binding experiments also suggest that IGF-II binds to another site on H295R cells. IGF-II binding to this second site was not inhibited by large amounts of IGF-I. This second site is obviously not an IGFBP as des(1-6)IGF-II (which does not bind to IGFBP) completely displaces 125I-IGF-II from H295R cells. This second site may rather represent the IGF-II M6P receptor, and we indeed detected this receptor in H295R cells by RT-PCR. Intact IGF-II M6P receptor has recently been demonstrated in normal adrenocortical tissues by Western immunoblotting (Weber et al. 1997) but the expression of the IGF-II M6P receptor in adrenocortical tumors has only been studied at the mRNA level; RT-PCR has been used to show that the gene is expressed both in normal and tumorous adrenals (Ilvesmäki et al. 1993).
IGFBP-2 has a 10-fold higher affinity for IGF-II than for IGF-I and is the IGFBP most frequently produced by malignant tumor cells (Reeve et al. 1992). Whether IGFBP-2 has an inhibitory or stimulatory effect on H295R cell proliferation remains to be demonstrated. Recently several indirect arguments indicate that IGFBP-2 may potentiate IGF action by focally concentrating the IGFs in the pericellular environment (Reeve et al. 1993, Arai et al. 1996, Russo et al. 1997) but the precise mechanism is still unknown. In the case of adrenocortical tumor cells, which are highly proliferative, IGFBP-2 may possibly have a
normal adrenal 100 bp ladder
day 4 11 15 18 20
447 bp -
type 1 IGF receptor
160 bp
type 2 IGF receptor
stimulatory role enhancing IGF-II action. Further studies are needed to elucidate the interaction between IGFBP-2 and IGF-II in the H295R model of adrenocortical tumors and to determine the
100
75
% inhibition
₸
50
T
₸
₸
L
キ
T
T
L
25
L
0
0
5 10 15 20 25 30
antibodies concentration (ug/ml)
localization of IGFBP-2 in the cellular and pericellular environment.
As stated above, malignant adrenocortical tumors frequently display imprinting abnormalities of the 11p15 region. Whether this occurred in the primitive adrenocortical tumor from which the H295R cell line is derived is not known. However, we found that the expression of two genes, both normally expressed from the maternal allele, H19 and p57KIP2, was abrogated (data not shown) suggesting that H295R cells share abnormalities of 11p15 imprinting with malignant adrenocortical tumors.
In conclusion, our observations show that the H295R cell line is a good model of adrenocortical tumorigenesis; the cells maintain steroid production (particularly androgen production) and, produce, like human malignant tumors, high amounts of IGF-II and IGFBP-2. We also demonstrate that IGF-II is directly involved in tumor proliferation and that its effects are mediated, at least in part, by the type 1 IGF receptor in an autocrine fashion.
ACKNOWLEDGEMENTS
This work was supported by Assistance Publique- Hôpitaux de Paris: Contrats de Recherche Clinique no. 940027 and 97134, by the University of Paris VI, Faculté Saint-Antoine (UPRES EA 1531) and by Association de Recherche contre le Cancer (no. 1364). A L was supported by a grant from the
Ministère de l’Education Nationale. We thank Dr W E Raincy for providing the NCI H295R cell line.
REFERENCES
Arai T, Busby W & Clemmons D 1996 Binding of Insulin-like growth factor (IGF) I or II to IGF-binding protein-2 enables it to bind to heparin and extracellular matrix. Endocrinology 137 4571-4575.
Backlin C, Rastad J, Skogseid B, Hellman P, Akerstrom G & Juhlin C 1995 Immunohistochemical expression of insulin-like growth factor 1 and its receptor in normal and neoplastic human adrenal cortex. Anticancer Research 15 2453-2460.
Binoux M, Lassarre C & Gourmelen M 1986 Specific assay for insulin-like growth factor (IGF) II using the IGF binding proteins extracted from human cerebrospinal fluid. Journal of Clinical Endocrinology and Metabolism 63 1151-1155. Boulle N, Logié A, Gicquel C, Perin L & Le Bouc Y 1998 Increased levels of Insulin-like growth factor-II (IGF-II) and IGF-binding protein-2 are associated with malignancy in sporadic adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 83 1713-1720. Chomczynski P & Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol- chloroform extraction. Analytical Biochemistry 162 156-159. Christofori G, Naik P & Hanahan D 1994 A second signal supplied by insulin-like growth factor II in oncogene- induced tumorigenesis. Nature 369 414-418.
Christofori G, Naik P & Hanahan D 1995 Deregulation of both imprinted and expressed alleles of the insulin-like growth factor 2 gene during ß-cell tumorigenesis. Nature Genetics 10 196-201.
De Chiara T, Robertson E & Efstratiadis A 1991 Parental imprinting of the mouse insulin-like growth factor II gene. Cell 64 849-859.
Fiet J, Gosling J, Soliman H, Galons H, Boudou P, Aubin P, Belanger A, Villette J, Julien R, Brérault J, Burthier J, Morineau G, Al Halnak A & Vexiau P 1994 Hirsutism and acne in women: coordinated radioimmunoassays for eight relevant plasma steroids. Clinical Chemistry 40 2296-2305.
Gazdar A, Oie H, Shackleton C, Chen T, Triche T, Myers C, Chrousos G, Brennan M, Stein C & La Rocca R 1990 Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Research 50 5488-5496.
Gicquel C, Bertagna X, Schneid H, Leblond-Francillard M, Luton J, Girard F & Le Bouc Y 1994 Rearrangements at 11p15 locus and overexpression of insulin like growth factor-II gene in sporadic adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 78 1444-1453. Gicquel C, Raffin-Sanson M, Gaston V, Bertagna X, Plouin P, Schlumberger M, Louvel A, Luton J & Le Bouc Y 1997 Structural and functional abnormalities at 11 p 15 are associated with the malignant phenotype in sporadic adrenocortical tumors. Study on a series of 82 tumors. Journal of Clinical Endocrinology and Metabolism 87 2559-2565.
Hardouin S, Gourmelen M, Noguiez P, Seurin D, Roghani M, Le Bouc Y, Povoa G, Merimee T, Hossenlopp P & Binoux M 1989 Molecular forms of serum insulin-like growth factor (IGF)-binding proteins in man: Relationships with growth hormone and IGFs and physiological significance. Journal of Clinical Endocrinology and Metabolism 69 1291-1301.
Journal of Molecular Endocrinology (1999) 23, 23-32
Hol F, Geurds M, Hamel B & Mariman E 1992 Improving the polymorphism content of the 3’ UTR of the human IGF2R gene. Human Molecular Genetics 1 347.
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. Annals of Biochemistry 154 138-143.
Ilvesmäki V, Kahri A, Miettinen P & Voutilainen R 1993 Insulin-like growth factors (IGFs) and their receptors in adrenal tumors: High IGF-II expression in functional adrenocortical carcinomas. Journal of Clinical Endocrinology and Metabolism 77 852-858.
Jones J & Clemmons D 1995 Insulin-like growth factors and their binding proteins: biological actions. Endocrine Reviews 16 3-34.
Kamio T, Shigematsu K, Kawai K & Tsuchiyama H 1991 Immunoreactivity and receptor expression of insulin-like growth factor I and insulin in human adrenal tumors. American Journal of Pathology 138 83-91.
Lalou C & Binoux M 1993 Evidence that limited proteolysis of insulin-like growth factor binding protein-3 (IGFBP-3) occurs in the normal state outside of the bloodstream. Regulatory Peptides 48 179-188.
Le Bouc Y, Noguiez P, Sondermeijer P, Dreyer D, Girard F & Binoux M 1987 A new 5’-non-coding region for human placental insulin-like growth factor II mRNA expression FEBS Letters 222 181-185.
Liu J, Kahri A, Heikkila P, Ilvesmaki V & Voutilainen R 1995 H19 and insulin-like growth factor-II gene expression in adrenal tumors and cultured adrenal cells. Journal of Clinical Endocrinology and Metabolism 80 492-496.
Ogawa O, Eccles M, Szeto J, McNoe L, Yun K, Maw M, Smith P & Reeve A 1993 Relaxation of insulin-like growth factor II gene imprinting implicated in Wilms’ tumour. Nature 362 749-751.
Rainey W, Bird I & Mason J 1994 NCI-H295 cell line: a pluripotent model for human adrenocortical studies. Molecular and Cellular Endocrinology 100 45-50.
Rainier S, Johnson L, Dobry C, Ping A, Grundy P & Feinberg A 1993 Relaxation of imprinted genes in human cancer. Nature 362 747-749.
Reeve J, Kirby L, Brinkman A, Hughes S, Schwander J & Bleehen N 1992 Insulin-like growth factor binding protein gene expression and protein production by human tumour cell lines. International Journal of Cancer 51 818-821.
Reeve J, Morgan J, Schwander J & Bleehen N 1993 Role for membrane and secreted insulin like growth factor binding protein-2 in the regulation of insulin like growth factor action in lung tumors. Cancer Research 53 4680-4685. Russo V, Bach L, Fosang A, Baker N & Werther G 1997 Insulin-like growth factor binding protein-2 binds to cell surface proteoglycans in the rat brain olfactory bulb. Endocrinology 138 4858-4867.
Staels B, Hum D & Miller W 1993 Regulation of steroidogenesis in NCI-H295 cells: a cellular model of the human fetal adrenal. Molecular Endocrinology 7 423-433. Weber M, Auernhammer C, Kiess W & Engelhardt D 1997 Insulin-like growth factor receptors in normal and tumorous adult human adrenocortical glands. European Journal of Endocrinology 136 296-303.
Werner H & LeRoith D 1996 The role of the insulin-like growth factor system in human cancer. Advances in Cancer Research 68 183-223.
REVISED MANUSCRIPT RECEIVED 23 December 1998