0013-7227/91/1283-1623$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society

Vol. 128, No. 3 Printed in U.S.A.

Insulin Stimulates Synthesis and Release of Human Chorionic Gonadotropin by Choriocarcinoma Cell Lines SONG-GUANG REN AND GLENN D. BRAUNSTEIN Department of Medicine Cedars-Sinai Medical Center- University of California School of Medicine, Los Angeles, California 90048

ABSTRACT. Recent studies have shown that insulin regulates placental lactogen, progesterone, and estrogen production from human trophoblast cells. This study was performed to examine whether insulin also regulates the production of hCG by this type of cell. After 24-36 h of preincubation, JEG-3 and JAR cells (2-3 x 105 cells/ml -well) or human term trophoblast cells (1 x 106 cells/ ml • well) were exposed to the test hormone in serum-free Dulbecco's Modified Eagle's Medium for 24-96 h. Secretion of hCG from JEG-3 cells was stimulated by human insulin, human proinsulin, or porcine insulin in a dose-dependent manner, with lowest effective doses of 6.7, 96, and 53 mg/L, respectively. Time-course studies showed that hCG secretion peaked at 7296 h with insulin exposure; in contrast, no decernable peak was seen without insulin in serum-free media. Exposure of JEG-3 cells for 24 h to 209 mg/liter insulin stimulated hCG synthesis, with 40 ± 3% more immunoreactive intracellular hCG (P < 0.05). Cells grown in the presence of insulin and [35S]methionine

I

NSULIN and other growth factors are important regulators of growth and function in many cell types. Insulin has been shown to potentiate PRL secretion and PRL mRNA levels in pituitary cells (1) and stimulate LH and FSH release by pituitary cells in vitro (2). Insulin and insulin-like growth factor-I (IGF-I) are also involved in the stimulation of synthesis and release of PRL from human decidual cells (3, 4) and regulation of 30-hydroxysteroid dehydrogenase activity and progesterone (5) as well as placental lactogen and estradiol production (6, 7) by human trophoblasts. Furthermore, epidermal growth factor (8-10) and fibroblast growth factor (11) have been documented to enhance hCG secretion by cultured human choriocarcinoma cells. Recently, during a study designed to evaluate the interaction of trophoblast and decidual cells, we found that a 6000- to 7000-dalton fraction of CMRL-1066 medium supplemented with 1 U/ml insulin and 10% fetal bovine serum significantly increased hCG secretion by Received September 17, 1990. Address all correspondence and requests for reprints to: Glenn D. Braunstein, M.D., Department of Medicine, B118, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, California 90048.

had 47 ± 21% more labeled intracellular hCG and 56 ± 13% more immunoprecipitable [35S]methionine-hCG secreted into the medium than the control cultures (P < 0.05). During this time period, human placental lactogen release and total trichloroacetice acid-precipitable [35S]methionine protein were not increased. The insulin-induced stimulation of hCG synthesis was inhibited by cycloheximide. Additionally, insulin did not significantly affect total intracellular protein during 24-96 h of incubation. Insulin also increased hCG release from JAR cells, but not from human term trophoblast cells. A mouse monoclonal antibody to the IGF-I receptor inhibited the stimulation of insulin in JEG-3 cells. We conclude that insulin stimulates the synthesis and secretion of hCG from JEG-3 cells and JAR cells, and that hCG regulation in choriocarcinoma cells differs from that in primary human placental trophoblast cells. The effect of insulin on JEG3 cells may be mediated in part through the insulin-like growth factor-I receptor. (Endocrinology 128: 1623-1629, 1991)

JEG-3 choriocarcinoma cells. Since insulin and/or IGFI fall within this mol wt range, and JEG-3 cells possess insulin (12) and IGF-I receptors (13), we hypothesized that insulin and/or IGF-I may stimulate hCG secretion by JEG-3 cells. After observing stimulation of hCG secretion by the JEG-3 cells, we evaluated the specificity of the response by examining the effect of insulin on the JAR choriocarcinoma cell line and freshly isolated human term trophoblast cells. Materials and Methods Cell preparation and cultures Two human choriocarcinoma cell lines, JEG-3 and JAR, were obtained from American Type Tissue Culture Collection (Rockville, MD). Human cytotrophoblast cells were isolated from term placental villus tissue through enzyme (trypsin and DNase, Sigma, St. Louis, MO) digestion and 5-70% Percoll (Sigma) gradient centrifugation, as described by Kliman et al. (14). Cell viability was greater than 95% using trypan blue exclusion. The cells (2-3 X 105 cells/ml • well for JEG-3 and JAR and 1 X 106 cells/ml • well for trophoblasts) were preincubated in 24multiwell culture plates with 16-mm well diameter (Costar, Cambridge, MA) containing Dulbecco's Modified Eagle's Me-

1623

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:01 For personal use only. No other uses without permission. . All rights reserved.

hCG STIMULATION BY INSULIN

1624

22

10. < 11.

dium (DMEM; Gibco, Grand Island, NY) supplemented with high glucose, 100 U/ml penicillin, 100 mg/ml streptomycin (Gibco; DMEM HG) and 10% fetal bovine serum (Gibco) in an atmosphere of 5% CO2-95% air at 37 C, then exposed to serumfree DMEM HG medium (control) or medium supplemented with hormones or other test substances for various times as described below. The experiments were terminated by removing medium. After the medium was removed, the cultured cells were washed with 0.01 M PBS (pH 7.4) and suspended in 1 M NaOH (for measuring protein) or in a lysis buffer (for assay of intracellular hormone) containing 0.02 M NaPO4, 0.2 M NaCl, 1% Triton X100, 0.2% sodium dodecyl sulfate, 1% Na deoxycholate, 0.05 M EDTA, and 0.8 raM phenylmethylsulfonylfluoride and subjected to freezing and thawing. The intracellular components of the lysed cells and the spent media were centrifuged, and the supernatants were stored at -20 C until hormone assay. In each experiment all treatments were tested in at least three wells, and each variable was tested in at least two separate experiments. Dose-response studies JEG-3 cells were preincubated for 36 h and then exposed for an additional 36 h to either serum-free DMEM HG medium (control) or the medium supplemented with varying concentrations of human insulin (0.015-417 mg/liter; recombinant DNA origin, Eli Lilly Co., Indianapolis, IN), human proinsulin (24192 mg/liter; Eli Lilly Co.), porcine insulin (26-417 mg/liter; Sigma), or IGF-I (10-500 Mg/liter; Amgen, Thousand Oaks, CA). The lowest effective dose was defined as the minimal dose tested at which hCG secretion by JEG-3 cells was significantly increased (P < 0.05) compared to the control. In another experiement JAR cells and primary human term trophoblast cells were preincubated for 36 h and then exposed for additional 36 h to control medium or the medium with varying concentration of human insulin (26-417 mg/liter). Time-course study JEG-3 cells were preincubated for 24 h and then exposed for 24-96 h to serum-free DMEM HG medium (control) or medium supplemented with human insulin at a dose of 209 mg/liter (experimental). The medium was exchanged every 24 h. The incubation of three wells from the control or experimental groups was terminated at 24, 48, 72, and 96 h of exposure, respectively. The hCG in the medium and that in the intracellular protein were each measured in a single assay. Inhibitory effect of cycloheximide To test whether the protein synthesis inhibitor cycloheximide inhibites insulin-stimulated hCG synthesis and release, JEG-3 cells were preincubated for 36 h and then exposed for an additional 24 h to either serum-free DMEM HG medium (control) or medium supplemented with 0.5 mM 8-bromo-cAMP (positive control; Sigma), 1 /uM cycloheximide (Sigma), 209 mg/ liter human insulin, or insulin plus cycloheximide. The hCG concentrations in the media or lysed cells were measured.

Endo • 1991 Voll28«No3

Total protein and hCG synthesis The synthesis of total protein and hCG was determined by measuring the incorporation of [35S]methionine into protein and hCG. After 36 h of preincubation, JEG-3 cells (5 X 105 cells/ml • well) exposed to serum- and methionine-free DMEM medium (control) or medium with human insulin (209 mg/liter) were incubated with 25 mCi/liter [35S]methionine (ICN, Irvine, CA) for 24 h. The culture medium was collected, and the cells were lysed using the method described above. The total protein present in the supernatant of lysed cells and medium was precipitated with 10% trichloroacetic acid (TCA) and collected on a glass microfiber filter (Whatman, Maidstone, England) by vacuum filtration. The intracellular hCG and the hCG in the medium were immunoprecipitated by incubation for 1 h at 37 C and for 16 h at 4 C with 25 ^1 polyclonal rabbit antihCG antibody (normal rabbit serum for control) and precipitation with sheep antirabbit -y-globulin and 6% polyethylene glycol. After centrifugation, the pellets were washed and recentrifuged in 0.01 M PBS buffer containing 5 mM methionine and dissolved in 1 M NaOH. The precipitated proteins and the hCG were assayed by liquid scintillation counting. Insulin and IGF-I binding Monolayer JEG-3 cells were washed with 0.01 M PBS buffer and incubated with 50,000 cpm [125I]IGF-I (SA, 252 /iCi//ig; Amersham, Arlington Heights, IL) or 30,000 cpm [125I]insulin (SA, 104 MCi//*g; New England Nuclear, Boston, MA) at 4 C for 16-20 h in serum-free DMEM HG medium with or without various concentrations of unlabeled IGF-I (10-700 /ug/liter) or insulin (0.16-834 mg/liter). The cells were washed with PBS buffer, pelleted by centrifugation, then dissolved in 1 M NaOH, and the radioactivity was counted. Nonspecific binding in the presence of 700 Mg/liter IGF-I (or 834 mg/liter insulin) was subtracted from each value to give IGF-I (or insulin) specific binding. Effect of anti-IGF-I receptor antiserum To determine the receptor through which insulin acts, we investigated the effect of an anti-IGF-I receptor antiserum on insulin-stimulated hCG release. After 24 h of preincubation, JEG-3 cells were cultured in serum-free DMEM HG medium for 24 h. After medium removal, the cells were exposed for 8 h to 0.5 ml medium containing 25 n\ normal mouse serum or mouse monoclonal antiserum to the IGF-I receptor (Oncogene Science, Inc., Manhasset, NY), followed by the addition of 0.5 ml control medium without insulin or medium with 209 mg/ liter insulin and an additional 30 h of incubation. The hCG concentrations in final 38-h cultured medium were measured. Measurements of hormones and protein hCG concentrations in the media or intracellular fractions, as well as human placental lactogen (hPL) in the media of selected experiments were measured by double antibody RIA methods, using reagents provided by the National Hormone Program, NIH (Bethesda, MD) (15, 16). The sensitivity of the assays were 1 ng/ml for both hCG and hPL. The intraassay coefficients of variation for the hCG and hPL RIAs were 7.7% and 11%, respectively, whereas the interassay coefficients of

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:01 For personal use only. No other uses without permission. . All rights reserved.

hCG STIMULATION BY INSULIN variation were 15.1% and 13.9%, respectively. All samples from a single experiment were run in the same assay. The protein contents of the cells and media were measured by the method of Bradford (17), using BSA as a standard and Bio-Rad protein reagents (Bio-Rad Laboratories, Richmond, CA). Concentrations of the hormones were expressed as nanograms per mg protein or nanograms per well. Statistical analysis All results are presented as the mean ± SEM. Student's t test with the Bonferroni correction for multiple observations was used for determining the statistical significance of the difference between control and experimental groups. P < 0.05 was considered significant.

Results As shown in Fig. 1, JEG-3 cells exposed for 36 h to the serum-free medium with varying concentrations of hu-

Human Insulin

1625

man insulin (0.015-417 mg/liter), human proinsulin (24192 mg/liter), or porcine insulin (26-417 mg/liter) demonstrated a dose-dependent stimulation of JEG-3 cell hCG secretion, with lowest effective doses of 6.7, 96, and 53 mg/liter for human insulin, human proinsulin, and porcine insulin, respectively. The stimulatory effect of human and porcine insulins on hCG release reached a peak at about a 300-350% increase above the control value. IGF-I at doses of 50-500 Mg/liter led to an approximately 50% increase in hCG secretion from JEG-3 cells, but this increase was not statistically significant and did not exhibit a dose-response relationship. The control values were 218 ± 17, 224 ± 11, 250 ± 14, and 119 ± 5 ng hCG/well • 36 h for human insulin, human proinsulin, porcine insulin, and IGF-I, respectively. JEG-3 cells exposed for 36 h to insulin (26-417 mg/liter) did not increase their hPL secretion compared to that of controls. The results of the time-course study are shown in Fig. 2. hCG secretion from the JEG-3 cells exposed to serumfree medium supplemented with 209 mg/liter insulin gradually increased over time of exposure, with peak values found at 72 h. hCG secretion from the cells *

.1

400

300

5

7T

•I

Insulin 209mg/liter

200

9 O

§

100

Control

o o o

Control 600 Insulin 209mg/liter 500

£ 9

Z 400 a 3 =

300

9

o 0.1 1 10 100 1000 CONCENTRATION OF NSULINS OR K3F-I (mg/litor) FIG. 1. The dose-response effect of insulin on hCG secretion by JEG3 cells. JEG-3 cells were exposed for 36 h to either serum-free DMEM HG medium (control) or medium supplemented with varying concentrations of human insulin (0.015-417 mg/liter), human proinsulin (24192 mg/liter), porcine insulin (26-417 mg/liter), or IGF-I (10-500 fig/ liter). The values are expressed as a percentage of the control value during 36-h exposure. Each point represents the mean ± SEM of triplicate cultures. *, P < 0.05 in comparison to control.

Baseline

24

48

72

96

Time (Hours)

FIG. 2. The time course of hCG release from JEG-3 cells and intracellular protein content after exposure to human insulin. JEG-3 cells were treated for 24-96 h with either serum-free DMEM HG medium (control) or medium with 209 mg/liter insulin. The medium was exchanged every 24 h. Each point represents the mean ± SEM hCG secreted during each 24 h (A) and intracellular protein content at each indicated incubation time (B) in triplicate cultures. *, P < 0.05 in comparison to control at the indicated time.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:01 For personal use only. No other uses without permission. . All rights reserved.

Endo'1991 Voll28«No3

hCG STIMULATION BY INSULIN

1626

exposed to serum-free medium alone, however, did not exhibit any peak during 96 h of exposure (Fig. 2A). The cells in either serum-free medium or medium with insulin continued to proliferate during the 96 h of incubation, as assessed by an increase in cell protein content in both groups (Fig. 2B). Human insulin also stimulated hCG synthesis. JEG-3 cells exposed for 24 h to 209 mg/liter insulin had a greater immunoreactive intracellular hCG content and more hCG in the medium compared to the controls (Table 1). The insulin-induced stimulation of hCG synthesis was completely inhibited by cycloheximide (Fig. 3). In addition, exposure of JEG-3 cells to 209 mg/liter insulin increased immunoprecipitable [35S]methionine-hCG by 56 ± 13% in medium and by 47 ± 21% in the intracellular pool (Table 1). Insulin at a dose of 209 mg/liter did not Significantly increase total [35S]methionone TCA-precipitable proteins during 24 h of exposure (Table 1) or total intracellular proteins during 96 h of exposure (Fig. 2B). Insulin also increased hCG release by another choriocarcinoma cell line, JAR, but did not stimulate hCG release from human term trophoblast cells in primary culture, as shown in Fig. 4. Binding studies confirmed that unlabeled insulin inhibited the binding of [125I]insulin to JEG-3 cells in a dose-dependent manner, as shown in Fig. 5. The halfmaximal displacement of labeled insulin occurred at 0.35 mg/liter. Unlabeled IGF-I also inhibited [125I]IGF-I binding to JEG-3 cells in a dose-dependent manner (Fig. 6), with half-maximal displacement at 10 /ug/liter. Unlabeled insulin was approximately l/100th as effective as IGF-I in competing for the IGF-I-binding sites (Fig. 6). When JEG-3 cells were exposed to 25 yul mouse anti-IGF-I receptor antiserum, insulin-stimulated hCG secretion by the cells was inhibited, as shown in Fig. 7. Discussion Our results demonstrate that insulin at pharmacological doses stimulates hCG synthesis and secretion by two

human choriocarcinoma cell lines, but not by human term trophoblasts. In JEG-3 cells, the insulin stimulatory effect on peptide hormone secretion is hCG specific, because hPL secretion is not changed by insulin, and the significant insulin-induced increase in hCG is much greater than the nonsignificant increase in TCA-precipitable protein in this cell line. A principal effect of insulin is the stimulation of hCG synthesis, since immunoprecipitable [35S]methionine-hCG is increased by insulin, and insulin-stimulated hCG synthesis and secretion are completely inhibited by cycloheximide. Although insulin has been reported to have growth stimulatory actions in vitro and in vivo (18), the lack of a difference in cellular protein in JEG-3 cells exposed for 24-96 h to medium either with or without insulin indicates that insulin-stimulated hCG production in JEG-3 cells is not secondary to enhanced mitogenesis. The secretory pattern of hCG from JEG-3 cells incubated for 96 h in serum-free medium with insulin differed from that of cells in serum-free medium without insulin, further supporting the concept that hCG production in JEG-3 cells is mediated in part by insulin. Our previous studies have shown that in a culture system where cells were incubated in DMEM medium with 10% fetal bovine serum, peak hCG release from primary trophoblast cells (19) and JEG-3 cells (unpublished data) occurred between 72-96 h of incubation, a pattern similar to that found in JEG-3 cells exposed to serum-free medium with insulin in the present study. A similar secretory pattern has been found when cAMP is added to JEG-3 cells (20). These observations suggest that either serum containing insulin and other growth factors or insulin itself is essential for JEG-3 cells to maintain the secretory pattern of hCG seen in the presence of serum. In contrast, Lobo and Bellino (21) found that the pattern of hCG secretion in primary trophoblast cells is not different in the presence or absence of serum. Thus, hCG production in

TABLE 1. Insulin enhances the synthesis of hCG in JEG-3 cells Cellular Immunoreactive hCG (ng/mg protein) Control Insulin TCA-precipitable [35S]methionine-protein (cpm x 105/well) Control Insulin Immunoprecipitable [35S]methionine-hCG (cpm x 103/well) Control Insulin

Secreted

Total

57 ± 6 80 ± 6 "

36 ± 2 102 ± 3°

93 ± 8 182 ± 9°

9.69 ± 1.08 10.58 ± 1.09

2.19 ± 0.27 2.79 ± 0.21

11.87 ± 1.35 13.38 ± 1.30

19.65 ± 1.23 28.82 ± 4.10°

2.14 ± 0.20 3.33 ± 0.19°

21.79 ± 1.43 32.15 ± 4.29"

JEG-3 cells exposed to sorum-free DMEM HG medium (control) or medium with insulin (209 mg/L) were incubated with [35S]methionine (25 jiCi/ml) for 24 h. The culture medium was collected, and the cells were homogenized. Levels of hCG in both cells and media were measured by RIA. [35S]Methionine cellular and secreted proteins were precipitated with TCA. [35S]Methionine hCG was isolated by immunoprecipitation. Each value represents the mean ± SEM of triplicate cultures. °P

Cyclohex(iuM)

FIG. 3. Inhibitory effect of cycloheximide on insulin-stimulated hCG synthesis and release. JEG-3 cells were exposed for 24 h to either serum-free DMEM HG medium (control) or medium supplemented with insulin (209 mg/liter), 8-bromo-cAMP (0.5 mM), cycloheximide (1 JUM), or insulin plus cycloheximide. The hCG in medium or lysed cells was measured. The values represent the mean ± SEM of six cultures. *, P < 0.05 in comparison to control for media or cells, respectively. 150T JZ CD CO

JAR Cells

Human Trophoblasts

£ o Q.

100-

c a) o

o O O O

Control 27

Insulin (mg/L)

53

105

209 417

Insulin (mg/L)

FlG. 4. hCG release from JAR cells and primary human trophoblasts in response to exposure to insulin. JAR cells or primary term trophoblasts were treated for 36 h with either serum-free DMEM HG medium (control) or medium with varying concentrations of human insulin (26417 mg/liter). Each bar represents the mean ± SEM of hCG in triplicate cultures. +,P< 0.05 in comparison to control.

response to stimulation by insulin in JEG-3 cells differs from that in human trophoblast cells. Other evidence supporting this concept is the failure of insulin to stimulate hCG secretion from human term trophoblasts, but not the JAR cell line and the lack of hPL secretion by insulin in the JEG-3 cells in the present study, in contrast to the stimulation of hPL by insulin in primary human trophoblast cells (6, 7). It is unlikely that the unresponsiveness of primary trophoblast hCG secretion to insulin is due to the use of trypsin, which may destroy insulin receptors on the cell surface, since trypsin also

.01

0.1

1

10

100

1000

INSULIN (mg/liter)

FIG. 5. Displacement of [125I]insulin binding to JEG-3 cells by unlabeled insulin. The cells were incubated with 30,000 cpm [125I]insulin for 16 h at 4 C in serum-free medium with or without various concentrations of unlabeled insulin (0.16-834 mg/liter). Nonspecific binding in the presence of 834 mg/liter insulin was subtracted from each value to give insulin specific binding.

was used in the passages of JEG-3 and JAR cells. These facts suggest that the insulin-regulated effect on hCG production is a property of choriocarcinoma cells and not primary trophoblasts. This is supported by our previous in vivo study which showed that circulating levels of hCG in pregnant diabetic women treated with insulin are not different from those in normal pregnancy (22). The lowest effective dose of insulin that induced a significant increase in hCG production in this study is much greater than insulin concentrations found under physiological conditions. The requirement of large quantities of insulin for a stimulatory effect in our experimental system is not due to excessive degradation of insulin in the culture medium, since 50% of the added insulin was detectable by RIA after 24 h of incubation (our unpublished data). Additionally, Oberbauer et al. (11) found that fibroblast growth factor-stimulated hCG production from JAR cells was enhanced by coadministration of insulin (2 mg/liter), but insulin alone at that dose did not have the same effect, supporting our finding that high doses of insulin are needed for an effect. Although JEG-3 cells contain insulin receptors on their surface (12,13), the stimulatory effect of insulin on hCG production by JEG-3 cells may be mediated through the IGF-I receptor. This possibility is suggested by the following facts. First, the effective range of insulin doses in regards to hCG stimulation is about 19 times greater than that required for displacement of labeled insulin

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:01 For personal use only. No other uses without permission. . All rights reserved.

1628

hCG STIMULATION BY INSULIN

Endo«1991 Voll28«No3

100-1

FIG. 6. Competitive inhibition of [125I] IGF-I binding to JEG-3 cells by varying concentrations of unlabeled IGF-I or insulin. Monolayer JEG-3 cells were incubated with 50,000 cpm [125I]IGF-I at 4 C for 20 h in serum-free DMEM HG medium with or without various concentrations of unlabeled IGF-I (10-700 Mg/liter; O- -O) or insulin (0.16-834 mg/liter; • - • ) . Nonspecific binding in the presence of 700 Mg/liter IGF-I was subtracted from each value to give specific binding. Each point represents mean of triplicate cultures.

50-

o 0.1

1

10

100

1000

Hormone Concentration (mg/L) 300

r 3 < o.oe 1

p < 0.0, ii

i

T

T

100

1

--

25 Ul mouse serum

25 AJI mouse serum + insulin 209 mg/liter

1

25JJI anti-IGF-l receptor antlserum + Insulin 209 mg/liter

FIG. 7. The effect of an anti-IGF-I receptor antiserum on insulinstimulated hCG release. After preincubation for 24 h in DMEM HG medium with 10% fetal bovine serum and 24 h in serum-free DMEM HG medium, JEG-3 cells were exposed for an additional 38 h to medium with 25 jul normal mouse serum, to medium with 25 jul normal mouse serum plus 209 mg/liter insulin, or to medium with 25 /il mouse monoclonal antiserum to IGF-I receptor plus 209 mg/liter insulin. The bars present the mean ± SEM hCG concentration of six wells during the final 38 h of incubation.

from the insulin receptor. Second, insulin has a low affinity for the IGF-I receptors on JEG-3 cells. Finally, a mouse monoclonal antibody to the IGF-I receptor inhibits the stimulative action of insulin on hCG production. The stimulatory effect on hCG by IGF-I did not occur when the dose of IGF-I was less than 500 ixg/liter (9,11), but some effect was seen at higher doses. We do not, however, exclude the possibility that this effect of insulin may be mediated through other mechanisms, such as interaction with the epidermal growth factor receptor, which may have a functional and structural

relationship to the insulin receptor (22), or through a nonreceptor-mediated intracellular action of insulin or one of its degradation products (24). In summary, this study provides evidence that insulin, but not IGF-I, stimulates hCG production by two choriocarcinoma cell lines. This effect is not a general property of all trophoblastic cells, since primary human placental term trophoblast cells are not stimulated by insulin.

Acknowledgments We thank Ms. Judith Seliktar and Ms. Elisa Gonzales for their technical assistance, Ms. Helene Zauderer for her secretarial help, and Dr. Mayer Davidson for his helpful suggestions. The provision of reagents by the National Hormone and Pituitary Program (Baltimore, MD) is gratefully acknowledged.

References 1. Prager D, Yamashita S, Melmed S 1988 Insulin regulates prolactin secretion and messenger ribonucleic acid levels in pituitary cells. Endocrinology 122:2946-2952 2. Adashi EY, Hsueh AJW, Yen SSC 1981 Insulin enhancement of luteinizing hormone and follicle-stimulating hormone release by cultured pituitary cells. Endocrinology 108:1441-1449 3. Thrailkill KM, Golander A, Underwood LE, Richards RG, Handwerger S 1989 Insulin stimulates the synthesis and release of prolactin from human decidual cells. Endocrinology 124:3010-3014 4. Thrailkill KM, Golander A, Underwood LE, Handwerger S 1988 Insulin-like growth factor I stimulates the synthesis and release of prolactin from human decidual cells. Endocrinology 123:2930-2934 5. Nestler JE 1989 Insulin and insulin-like growth factor-I stimulate the 3/3-hydroxysteroid dehydrogenase activity of human placental cytotrophoblasts. Endocrinology 125:2127-2133 6. Hochberg Z, Perlman R, Brandes JM, Benderli A 1983 Insulin regulates placental lactogen and estradiol secretion by cultured human term trophoblast. J Clin Endocrinol Metab 57:1311-1313 7. Bhaumick B, Dawson EP, Bala RM 1987 The effects of insulinlike growth factor-I and insulin on placental lactogen production by human term placental explants. Biochem Biophys Res Commun 144:674-682 8. Benveniste R, Speeg Jr KV, Carpenter G, Cohen S, Linder J, Rabinowitz D 1978 Epidermal growth factor stimulates secretion of human chorionic gonadotropin by cultured human choriocarcinoma cells. J Clin Endocrinol Metab 46:169-172 9. Ritvos O, Jalkanen J, Pekonen F, Stenman U-H, Ranta T 1988 Epidermal growth factor but not insulin-like growth factor-I potentiates adenosine 3',5'-monophosphate-mediated chorionic gonadotropin secretion by cultured human choriocarcinoma cells.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:01 For personal use only. No other uses without permission. . All rights reserved.

hCG STIMULATION BY INSULIN Endocrinology 123:859-865 10. Ilekis J, Benveniste R 1985 Effects of epidermal growth factor, phorbol myristate acetate, and arachidonic acid on choriogonadotropin secretion by cultured human choriocarcinoma cells. Endocrinology 116:2400-2409 11. Oberbauer AM, Linkhart TA, Mohan S, Longo LD 1988 Fibroblast growth factor enhances human chorionic gonadotropin synthesis independent of mitogenic stimulation in JAR choriocarcinoma cells. Endocrinology 123:2696-2700 12. Deal CL, Guyda HJ 1983 Insulin receptors of human term placental cells and choriocarcinoma (JEG-3) cells: characteristics and regulation. Endocrinology 112:1512-1523 13. Ritvos O, Rutanen E-M, Pekonen F, Jalkanen J, Suikkari A-M, Ranta T 1988 Characterization of functional type I insulin-like growth factor receptors from human choriocarcinoma cells. Endocrinology 122:395-401 14. Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss III JF 1986 Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology 118:15671582 15. Braunstein GD, Karow WG, Gentry WC, Rasor J, Wade ME 1978 First-trimester chorionic gonadotropin measurements as an aid in the diagnosis of early pregnancy disorders. Am J Obstet Gynecol 131:25-32 16. Braunstein GD, Rasor J, Engvall E, Wade ME 1980 Interrelationships of human chorionic gonadotropin, human placental lactogen,

17. 18. 19. 20.

21.

22.

23. 24.

1629

and pregnancy-specific j8i-glycoprotein throughout normal human gestation. Am J Obstet Gynecol 138:1205-1213 Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248-254 Straus DS 1984 Growth-stimulatory actions of insulin in vitro and in vivo. Endocr Rev 5:356-368 Kato Y, Braunstein GD 1989 Discordant secretion of placental protein hormones in differentiating trophoblasts in vitro. J Clin Endocrinol Metab 68:814-819 Chou JY 1980 Regulation of human chorionic gonadotropin synthesis in placenta, choriocarcinoma, and non-placental tumors. In: Segal SJ (ed) Chorionic Gonadotropin. Plenum Press, New York, pp 317-339 Lobo JO, Bellino FL 1989 Estrogen synthetase (aromatase) activity in primary culture of human term placental cells: effects of cell preparation, growth medium, and serum on adenosine 3',5'-monophosphate response. J Clin Endocrinol Metab 69:868-874 Braunstein GD, Mills JL, Reed GF, Jovanovic LG, Holmes LW, Aarons J, Simpson JL, NICHHD-Diabetes in Early Pregnancy Study Group 1989 Comparison of serum placental protein hormone levels in diabetic and normal pregnancy. J Clin Endocrinol Metab 68:3-8 Rosen OM 1987 After insulin binds. Science 237:1452-1458 Duckworth WC, Kitabchi AE 1981 Insulin metabolism and degradation. Endocr Rev 2:210-233

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 05:01 For personal use only. No other uses without permission. . All rights reserved.

Insulin stimulates synthesis and release of human chorionic gonadotropin by choriocarcinoma cell lines.

Recent studies have shown that insulin regulates placental lactogen, progesterone, and estrogen production from human trophoblast cells. This study wa...
857KB Sizes 0 Downloads 0 Views