Mitogenic effect of basic fibroblast growth factor and estradiol on cultured human myometrial and leiomyoma cells

Mitogenic effect of basic fibroblast growth factor and estradiol on cultured human myometrial and leiomyoma cells Phillip George

N. Rauk, MD,” Urvashi Surti, Michalopoulos, MD, PhD

Pittsburgh,

PhD,”

c James M. Roberts,

MD,” and

Pennsylvania

OBJECTIVE: We compared the mitogenic effect of basic fibroblast growth factor with and without estradiol on myometrial and leiomyometrial cells. STUDY DESIGN: The mitogenic effect of basic fibroblast growth factor on myometrial cells was measured by thymidine incorporation and cell count. The mitogenic effect of basic fibroblast growth factor with and without estradiol as measured by thymidine incorporation was compared between myometrial and leiomyometrial cells. RESULTS: Both human myometrial and leiomyometrial cells showed significant @ = 0.004 and p = 0.001, respectively), dose-dependent incorporation of thymidine in response to basic fibroblast growth factor. Leiomyometrial cells showed significantly @ = 0.04) less thymidine incorporation compared with matched normal myometrial cells. The addition of estradiol with basic fibroblast growth factor did not result in a further increase in thymidine incorporation. CONCLUSIONS: Both myometrial and leiomyometrial cells respond to basic fibroblast growth factor with increased thymidine incorporation; however leiomyometrial cells are less responsive than are matched normal myometrial cells. The addition of estradiol is not synergistic with basic fibroblast growth factor. (AM J OEISTET GYNECOL 1995;173:571-7.)

Key words:

Basic fibroblast

growth

factor, leiomyoma,

Uterine leiomyomas are benign smooth muscle tumors occurring in as many as 30% of women > 35 years old. Although these tumors are asymptomatic in many women, they are a frequent cause of abnormal bleeding, pelvic pain, reduced fertility, and fetal wastage. As a result, the presence of leiomyomas is a leading cause for hysterectomy. The majority of patients have multiple well-circumscribed leiomyomas, and each leiomyoma is thought to be clonal, arising independently from a single initiated smooth muscle cell.’ The nature of the initial event is unknown. There is evidence, however, that implicates estrogen as a factor in leiomyoma development. Leiomyomas grow during the reproductive years, increase in size during pregnancy, and regress during menopause. Treatment with gonadotropin-reFrom the Division of Maternal-Fetal Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences,” and the Departments of Human Genetics” and Pathology,“. University of Pittsburgh. Supported by the Magee- Womens Hospital Foundation Research Fund. Presented at the Fortieth Annual Meeting of the Society for Gynecologic Investigation, Toronto, Ontario, Canada, March 31-April 3, 1993. Received for publication October 11, 1994; revised November 7, 1994; accepted December 15, 1994. Reprint requests: Phillip N. Rauk, MD, Department of Obstetrics, Gynecology, and Reproductive Sciences, Maeee- Womens Hospital, 360 Haget St., Pit&burgh, PA 15213. v Copyright 0 1995 by Mosby-Year Book, Inc. 0002-9378/95 $3.00 + 0 6/l/62763

estradiol leasing hormone analogs, which reduces serum estradiol concentrations, also leads to a reduction in the size of leiomyomas; however, reenlargement occurs after therapy is discontinued.’ Tissue concentrations of estrogen receptors are increased in leiomyomas compared with normal myometrium.3 Although these findings support a role for estrogen on the growth of leiomyomas in vivo, there has been little in vitro evidence to support estrogen as the sole mediator of leiomyoma growth. The original concept of an indirect estrogen effect mediated hormonally by circulating “estromedins” has not been entirely supported experimentally. Currently it is postulated that the effect of estrogen is exerted through activation of paracrineautocrine factors that modify growth. Several growth factors, specifically epidermal growth factor (EGF) and insulin-like growth factor I (IGF-I), have been implicated as mediators of leiomyoma growth in vivo.4 Another growth factor found in abundance in myometrium is basic fibroblast growth factor (FGF).” Basic FGF plays a central role in modulating cell proliferation, motility, differentiation, and survival in many mesodermally derived tissues.6 Basic FGF has been demonstrated in both benign and malignant tumors of the central nervous system and in human endometrial adenocarcinoma cell lines, where its production is estrogen regulated.‘. * As an angiogenic and mitogenic agent, basic FGF is believed to play an important role in 571

572

Rauk

Table

I. Clinical

Patient No.

and pathologic

characteristics

of study subjects

Age (Yd

Race

G/P

Tissue

1 2

50 50

w w

212 513

M M

3 4

38 45

A w

o/o l/l

M MIL

Menorrhagia Menorrhagia

5

63

W

l/l

M/L

Adnexal

6

78

W

513

M/L

7 8

40 46

w W

313 212

M/L

Uterine mass, menopausal, chronic hypertension Menorrhagia Menorrhagia

P, parity;

W, white;

A, Asian;

through

the maintenance

G, Gravidity;

tumor

.4ugust 1995 Am j Obstet Gynecol

et al.

growth

M/L

Medical

history

Pathologic

Enlarged uterus Menorrhagia

mass,

M, myometrium;

of tumor

blood

supply. In the current study we examined the mitogenic properties of basic FGF on human uterine myometrial and leiomyoma cells in culture by thymidine incorporation. We confirmed that increases in thymidine incorporation by normal myometrium also corresponded to increases in cell number. In addition, we examined the effect of estradiol on the mitogenic properties of basic FGF in both cell types by thymidine incorporation. Our data show that basic FGF is mitogenic for both human uterine myometrial and leiomyoma cells in vitro and that leiomyoma cells are less responsive to the mitogenie effect of basic FGF. Estradiol exhibits no synergism with the mitogenic effect of basic FGF in vitro.

Material and methods This study was approved by the Magee-Womens Hospital Human Use and Experimentation Committee. Culture of uterine smooth muscle cells. Samples of human uterine myometrium and leiomyoma from the same patient were obtained at hysterectomy. Tissue samples were placed in modified Hanks’ balanced salt solution (500 ml of calciumand magnesium-free Hanks’ balanced salt solution with 5 ml of heparin [lo00 U/ml], 5 ml of gentamicin [50 mg/ml], and 5 ml of penicillin G-streptomycin [ 10,000 U/ml and 10,000 pg/ml]) and kept at 4” C before processing. By use of sterile technique, the tissue samples were minced, washed with Earle’s balanced salt solution, and placed in a 15 ml conical tube and centrifuged at 5OOg for 5 minutes. The pellet was resuspended in 5 ml of 0.125% trypsin solution and incubated at 37” C for 15 minutes. Tissue was centrifuged at 500g for 5 minutes and resuspended in 5 ml of collagenase type II solution (Worthington, Freehold, N.J.), 8 mg in 12.5 ml of Earle’s balanced salt solution. The suspension was incubated at 37” C for 2 to 3 hours with occasional pipetting. The suspension was then passed through fine-mesh gauze, and individual cells were collected by

menopausal

diagnosis

Leiomyoma 19 cm, menstrual endometrium Leiomyoma 3.5 cm, adenomyosis, menstrual endometrium Leiomyoma 3.5 cm, proliferative endometrium Cellular leiomyoma 11 x 9 cm, adenomyosis, secretory endometrium Multiple leiomyoma, largest leiomyoma 2 cm, adenomyosis, inactive endometrium Leiomyoma 8 cm, proliferative endometrium Leiomyoma Leiomyoma

7 cm, proliferative 26 cm, proliferative

endometrium endometrium

L, leiomyoma.

centrifugation at 5009 for 5 minutes. The cells were washed twice with 10% minimal essential medium with nonessential amino acids, sodium bicarbonate (26 mmol/L), pyruvate (1 mmol/L), gentamicin (50 pg/ml), penicillin G (100 U/ml), streptomycin (100 kg/ml), L-glutamine (2 mmoliL), and charcoal-stripped fetal calf serum (10% [vol/vol]). The cells were then plated at 5 x 10’ cells in 25 cm’ flasks and maintained at 37” C in humidified 5% carbon dioxide. The medium was changed every 2 to 3 days, and the cells were subcultured at confluence after 5 to 7 days. The cells were identified at each passage as smooth muscle cells by means immunohistochemical staining with a monoclonal anti-c+smooth muscle actin antibody (Sigma, St. Louis). Thymidine incorporation. Cells were plated in 12well plates (3.8 cm”/well) at 2 X lo4 cells per well in 10% minimal essential medium with nonessential amino acids. After the cells had reached 75% confluence, the medium was changed to 1% minimal essential medium with nonessential amino acids for 48 hours to synchronize the cells in G,. After 48 hours either 1% minimal essential medium with nonessential amino acids alone or 1% medium with 17P-estradiol (10m8 mol/L) was added for an additional 24 hours. Basic FGF at concentrations from 0.1 to 5.0 ngiml was then added. Preliminary experiments demonstrated minimal thymidine incorporation in the first 16 hours and peak thymidine incorporation between 18 and 24 hours after the addition of basic FGF. After 18 hours thymidine 2 @.X/ml was added for 6 hours. The medium was then aspirated and the cells fixed with 5% trichloroacetic acid and washed with water. The trichloroacetic acid precipitable material was then resuspended in 0.33N sodium hydroxide. An aliquot was added to scintillation cocktail and counted on a scintillation counter. Cell count. Cells cultured from the normal myometrium

of patients

4, 5, 6, 7, and

8 were

plated

in six-well

plates (9.6 cm’iwell) at 1 x 10” cells per well in 10% minimal essential medium with nonessential amino

Volume 1’73, Number Am J Obstet Gynecol

Rauk et al.

2

T

T

-

‘T ’

ii Basic FGF Concentration

573

T

5.00

(ng/mL)

Fig. 1. Effect of basic FGF with and without 17p-estradiol on thymidine uptake in human uterine myocytes. Human uterine myocytes were treated with basic FGF at different concentrations with (solid bars) and without (open bars) 17P-estradiol (10 nmol/L). Results are mean t SEM (n = 8). Cells treated with basic FGF alone and in combination with 17P-estradiol showed significant dosedependent increase in thymidine uptake by analysis of variance Cp = 0.002 and p = 0.004, respectively). Cells treated with basic FGF alone and in combination with 17P-estradiol were compared by analysis of variance; no significant difference was observed in thymidine uptake.

Time (days) Fig. 2. Effect of basic FGF on human uterine myocyte cell number over uterine myocytes were treated with saline solution (open squares) or 1 .O ngiml cell number was counted at 0, 3, 5, and 7 days. Results are means f SEM and without basic FGF were significant Cp = 0.0001) at all times.

acids.

After

medium with

the

was

cells

had

changed

nonessential

to

amino

acids

After basic

75% confluence,

reached 1% minimal

nize the cells or 1% medium

in G,. with

the wells. The 5, and 7 days.

cells The

were counted medium was

analysis.

Statistical

for

essential

72 hours

72 hours either FGF, 1 ngiml,

the medium

in triplicate wells at 3, changed at 3, 5, and 7 analysis

two-way

dent

repeated

analysis

of variance

and

the

Stu-

t test.

to synchro1% medium was added to

days. Statistical

by

time in culture. Human basic FGF (closed squares); (n = 5). Differences with

was performed

Results Subjects. istics

of the

Relevant study

Both myometrial cessfully cultured

the leiomyomas

clinical subjects

and are

pathologic

summarized

characterin Table

I.

and leiomyoma tissue could be sucin five of the eight subjects. Cells from

of patients

1, 2, and 3 did not grow in

574

Rauk

August 1995 Am J Obstet Gym01

et al.

30000 g m

8

C

? EB

20000

22 m s 3

x

J TT II

10000

% .w E h c

0

0.00

0.10

0.50

0.75

1.00

Basic FGF Concentration

2.50

5.00

(ng/mL)

Fig. 3. Effect of basic FGF on thymidine uptake in human uterine myoeytes and leiomyocytes. Human uterine myocytes (open bars) and leiomyocytes (solid bars) were treated with basic FGF at different concentrations. Results are mean & SEM (n = 5). Analysis of variance showed a significant difference between myocytes and leiomyocytes (p = 0.04). Both myocytes and leiomyocytes showed a significant dose-dependent increase in thymidine uptake by analysis of variance (fi < 0.05). sufficient

quantity

for

the

study

of

thymidine

incorpo-

ration.

Effect of basic and cell number. showed poration centrations rig/ml

FGF

a significant of

on thymidine

Human (p = 0.004)

thymidine of basic

compared

in FGF,

with

incorporation

uterine

dose-dependent

response with

a peak

unstimulated

myometrial to

cells incor-

increasing

con-

of 3.5-fold

at 1.0

cells

(Fig.

1). The

addition of basic FGF to normal myometrial cells from patients 4, 5, 6, 7, and 8 also resulted in a significant (p < 0.001) increase in cell number at 3, 5, and 7 days of basic FGF exposure compared with controls (Fig. 2). The increase in cell number at 5 days was 2.8-fold compared with control. Comparison of the effect of basic FGF on thymidine incorporation between myometrial and leiomyoma cells. The response of normal myometrial cells and leiomyoma cells was compared in patients 4, 5, 6, 7, and 8, where both tissues were cultured. Fig. 3 demonstrates significantly ($ = 0.04) less thymidine incorporation by basic FGF at all concentrations in leiomyoma cells compared with normal myometrial cells. Leiomyoma cells, however, demonstrated a significant (p = 0.001) dose-dependent incorporation of thymidine in response to increasing concentrations of basic FGF. Effect of combined estradiol and basic FGF on thymidine incorporation in myometrial and leiomyoma cells. Neither in normal myometrial cells (Fig. 1) nor in leiomyoma cells (Fig. 4) did the addition of 10 nmol/L

17B-estradiol

result

in an increase

in thymidine

incorporation at any concentration of basic FGF or a shift in the dose-response curve to basic FGF. In myo-

metrial cells the concentration of basic FGF at halfmaximal stimulation was 0.28 -+ 0.14 rig/ml with 17@ estradiol and basic FGF compared with 0.32 +- 0.15 rig/ml in cells treated with basic FGF alone. In leiomyoma cells the concentration at half-maximal stimulation was 0.27 2 0.16 rig/ml with 17@estradiol and basic FGF compared with 0.30 ? 0.20 rig/ml with basic FGF alone. Comparison of concentration at halfmaximal stimulation by Student t test showed no significant difference between treatments in either cell type.

Comment We demonstrated that basic FGF is mitogenic for human uterine myometrial cells grown in culture. Our initial interest in basic FGF as a potential mitogen for uterine muscle arose from a single report demonstrating specific binding of basic FGF antibody to human uterine smooth muscle cells,’ The intensity of staining was similar to other tissues rich in basic FGF, including liver and heart. Basic FGF is mitogenic for cells grown in culture from almost all other mesenchymal-derived tissues, including vascular muscle, gastrointestinal muscle, cardiac muscle, and neuroectoderm.6 Therefore the findings of our study support a potential role for basic FGF in uterine muscle cell growth. Basic FGF has been implicated in the uterine remodeling that occurs at the time of implantation in the mouse pregnancy.* In contradiction, however, Beck and Garner” by use of rat uterine myometrial cells in primary culture, did not show an increase in thymidine uptake by basic FGFstimulated cells. Other than the elimination of insulin from

the

culture

media

and

species

difference,

the

Volume 173, Number Am J Obstet Gynecol

2

Rauk

et

al. 575

25000 G = 8 v)0 e E

20000

3

15000

s s 3

B ‘ii.m

E h is

1 r

IT TT~1

10000

TT

5000

0

1 0.00

Basic FGF Concentration

(ng/mL)

Fig. 4. Effect of basic FGF with and without 17@estradiol on thymidine uptake in human uterine leiomyocytes. Human uterine leiomyocytes were treated with basic FGF at different concentrations with (solid bars) and without (open bars) 17P-estradiol (10 nmol/L). Results are mean f SEM (n = 5). Cells treated with basic FGF alone and in combination with 17P-estradiol showed significant dose-dependent increase in thymidine uptake by analysis of variance (p c 0.05). Cells treated with basic FGF alone and in combination with 17P-estradiol were compared by analysis of variance; no significant differences was observed in thymidine uptake.

methods for culture and thymidine uptake were similar to our methods. The dose-response curve to basic FGF in our study is similar to that reported for a rat lung fibroblast cell line and primordial germ cells.“, I2 The maximum stimulation of deoxyribonucleic acid synthesis as measured by thymidine incorporation was at a concentration of basic FGF of 1 rig/ml, or 56 pmol/L, with half-maximal stimulation at approximately 0.30 rig/ml, or 18 pmol/L. Receptor-.binding studies have been performed in several cultured cells with binding affinity ranging from 20 to 200 pmol/L.’ Although we did not measure basic FGF binding to the myometrial cells, our dose-response data are consistent with the binding affinities reported in the literature. We can also be confident that our cells are muscle cells, as demonstrated by specific immunocytochemical staining with muscle-specific actin antibody. The human leiomyoma cells grown in culture demonstrated the same dose-response curve to basic FGF as did the myometrial cells. There was no difference in the concentrations of basic FGF resulting in either maximal or half-maximal stimulation. Our finding of lower thymidine incorporation at any basic FGF concentration in leiomyoma compared with matched myometrial cells is consistent with reports of other growth factors.‘3 Fayed et a1.13 found less thymidine incorporation by human leiomyoma cells in response to EGF, insulin alone, and platelet-derived growth factor than did matched myometrial cells. Other investigators also demonstrate less proliferation of leiomyoma cells under standard serum-

rich culture conditions.‘4 In general, in our study leiomyoma cells reached confluence at the same time as myometrial cells when plated at the same density and the same serum conditions. We did not measure the basic FGF receptor number on leiomyoma cells compared with normal myometrial cells. However, the report of Fayed et al.” did not indicate a correlation between receptor number and mitogenic response. Receptor concentration for platelet-derived growth factor was higher, insulin the same, and EGF lower in leiomyometrial cells in spite of an observed decrease in proliferative response to all three growth factors.13 Factors involved in intracellular signaling consequent to receptor binding may be altered in leiomyoma cells, accounting for this observed inconsistency. The lack of enhanced proliferation in response to several growth factors in vitro is inconsistent with the enhanced growth observed in vivo. Several reports show enhanced binding of IGF-I to leiomyoma cells, although no in vitro mitogenic studies have been performed.6* I5 Basic FGF may still play a role in leiomyoma growth in vivo through its angiogenic action and maintenance of tumor blood supply. Both immunocytochemical and in situ hybridization studies may delineate the cell types and differences in localization of basic FGF to tumor tissues and to fibroblasts and vascular elements. Our studies failed to show enhanced mitogenesis with combined estradiol and basic FGF. The action of estradiol on human myometrial basic FGF production and receptor concentration has not been investigated. Basic

576

Rauk

et al.

~4ugLlst 1995 Gynecol

Am J Obstet

FGF production is regulated by estradiol in other systems. Endometrial cell lines, when stimulated by estradiol, produce more basic FGF, but effects on growth were not investigated.’ Enhanced proliferation of the human ovarian cell line A2780 occurs with addition of basic FGF in combination with estradiol.‘6 Whether an autocrine mechanism of basic FGF action in these cells occurs has not been investigated. The stromal cells of rat prostate exposed to estradiol in vivo increase basic FGF messenger ribonucleic acid.” Our inability to demonstrate any estrogen synergism on basic FGF in myometrial cells may be explained by the loss of estrogen receptors. Eros1 et al.” showed a progressive decline in estrogen receptor concentration in cultured cells. By day 8 of secondary culture, at a time similar to that at which we performed our studies (5 to 10 days), estrogen receptor concentrations decreased by 75%. An alternative explanation would be increased endogenous basic FGF production in response to estrogen with subsequent down-regulation of the response to exogenous basic FGF. We did not measure basic FGF in the media after estradiol exposure; however, the mitogenic response to basic FGF combined with estradiol was not blunted compared with that of basic FGF alone. The regulation of other growth factors and growth factor receptors by estrogen has been well studied in myometrium. In vivo and in vitro studies have shown an increase in EGF protein and messenger ribonucleic acid and in IGF-I protein and messenger ribonucleic acid in both mouse and rat uterus in response to estrogen treatment.lg. TO Wbether this increase translates into increased mitogenic activity by these growth factors has not been studied. A system for guaranteeing preserved estrogen responsiveness in cultured cells would be helpful in answering these questions. There was also no synergism of basic FGF and estradiol in cultured leiomyoma cells. Although several studies show an increase in estrogen receptors in leiomyoma tissues compared with normal myometrium, we expect a similar loss of receptors in cultured leiomyoma cells, as with myometrial cells.” Several studies in human subjects suggest that both EGF and IGF-I protein synthesis and receptor concentrations are regulated by estrogen in vivo in leiomyomas. EGF and IGF-I receptor concentration decreases in leiomyomas from women rendered hypoestrogenic with gonadotropin-releasing hormone agonist.“. z3 IGF-I and IGF-I binding protein gene expression is also highest during the late proliferative phase of the menstrual cycle and absent during gonadotropin-releasing hormone administration.24 Leiomyomas grow in response to increased estradiol levels in vivo, yet no studies demonstrate an ability of estrogen to increase thymidine incorporation of cultured leiomyoma cells. This is true even in early primary cul-

tures where estrogen receptor concentrations remain highest. This lack of responsiveness in pure myocyte cultures strongly implicates estrogen paracrine effects. In conclusion, we demonstrated that both myometrial cells and leiomyoma cells respond to basic FGF with increased thymidine incorporation. The addition of estradiol is not synergistic with basic FGF. Leiomyoma cells are less responsive to basic FGF than are matched normal myometrial cells. The enhanced growth of leiomyomas over normal myometrium in vivo remains unexplained. The lack of enhanced proliferation by basic FGF in our cell culture system does not rule out other roles for basic FGF in the growth of leiomyomas. The action of basic FGF as an angiogenic factor may still play a role and requires further investigation. The role of other growth factors and the role of oncogene expression in leiomyoma cells also requires further study. Several studies report consistent chromosomal arrangements in 40% to 50% of leiomyomas.” Abnormalities involving translocation of 12q and 14q, deletion in chromosome 7, and alteration in chromosome 1 are the most common. The identification of novel tumor regulatory genes and growth factors and their receptors in the regions of these chromosome alterations is currently an area of active research. Because leiomyomas are thought to be clonal in origin, arising from a single cell, the role of apoptosis regulation, programmed cell death, in leiomyoma cells compared with normal myometrium may also give clues into the growth advantage of these cells in vivo. Data in this and other studies have yet to elucidate this estrogen effect, although many of these studies are confounded by the potential loss of estrogen responsiveness in vitro. We thank Jie Hu, PhD, and William technical assistance with the myometrial cell culture.

Bowen for their and leiomyoma

REFERENCES

1. Buttram VC, Reiter RC. Uterine leiomyomata: etiology, symptomatology, and management. Fern1 Steril 1981;36: 433-45. 2. West CP, Lumsden MA, Lawson S, Williamson J, Baird DT. Shrinkage of uterine fibroids during therapy with goserelin (Zoladex): a luteinizing hormone-releasing hormone agonist administered as a monthly subcutaneous depot. Fertil Steril 1987;48:45-5 1. 3. Otsuka H, Shinohara M, Kashimura M, Yoshida K, Okamura Y. A comparative studv of the estrogen receptor ratio in myometrium and uterine leiomyomas. Int J’Gynecol Obstet 1989;29:189-94. 4. Tommola P, Pekonen F, Rutanen EM. Binding of epiderma1 growth factor and insulin-like growth factor I in human myometrium and leiomyomata. Obstet Gynecol 1989;74:658-62. 5. Cordon-Card0 C, Vlodavsky I, Haimovitz-Friedman A, Hicklin D, Fuks Z. Exnression of basic fibroblast growth factor in normal hum& tissues. Lab Invest 1990;&:83240. 6. Rifkin DB, Moscatelli D. Recent developments in the cell

Volume 173, Number Am J Obstet Gym01

7.

8.

9.

10.

11.

12.

13.

14.

15.

Rauk

2

biology of basic fibroblast growth factor. J Cell Biol 1989; 109:1-6. Takahashi JA, Mori H, Fukumoto M, et al. Gene expression of fibroblast growth factors in human gliomas and meningiomas: demonstration of cellular source of basic fibroblast growth factor mRNA and peptide in tumor tissues. Proc Nat1 Acad Sci U S A 1990;87: 5710-4. Presta M. Sex hormones modulate the synthesis of basic Iibroblast growth factor in human endometrial adenocarcinema cells: implications for the neovascularization of normal and neoplastic endometrium. J Cell Physiol 1988; 137:593-7. Carlone DL, Rider V. Embryonic modulation of basic libroblast growth factor in the rat uterus. Biol Reprod 1993;49:653-65. Beck CA, Garner CW. Stimulation of DNA synthesis in rat uterine cells by growth factors and uterine’extracts. Mol Cell Endocrinol 1992:84:109-18. Paris S, Pouyssegur ‘J. Mitogenic effects of fibroblast growth factors in cultured fibroblasts. Ann N Y Acad Sci 1991;628:139-48. Resnick JL, Bixler LS, Cheng L, Donovan PJ. Long-term proliferation of mouse primordial germ cells in culture. Nature 1992;359:550-1. Fayed YM, Tsibris JCM, Langenberg PW, Robertson AL. Human uterine leiomyoma cells: binding and growth responses to epidermal growth factor, platelet-derived growth factor, and insulin. Lab Invest 1989;60:30-7. Cramer SF, Robertson AL, Ziats NP, Pearson OH. Growth potential of human uterine leiomyomas: some in vitro observations and their implications. Obstet Gynecol 1985; 66:36-41. Gloudemans T, Prinsen I, Van Unnik JAM, Lips CJM, Otter WD, Sussenbach JS. Insulin-like growth factor gene expression in human smooth muscle tumors. Cancer Res 1990;50:6689-95.

et al.

577

16. Speirs V, Jenkins S, White MC. Growth factor regulation of 17B-hydroxysteroid dehydrogenase activity in a human ovarian cell line: modulation by 17B-estradiol. Anticancer Res 1993;13:1399-404. 17. Bather M, Rausch U, Goebel HW, Polzar B, Mannherz HG, Aumuller G. II: Stromal and epithelial cells from rat ventral prostate during androgen deprivation and estrogen treatment-regulation of transcription, Exp Clin Endocrinol 1993;101:78-86. 18. Krosl J, Breskvar K, Hudnik-Plevnik T. Prolonged cultivation of rat uterine cells with preserved estrogen responsiveness. J Steroid Biochem 1989;33: 189-94. 19. Lingham RB, Stance1 GM, Loose-Mitchell DS. Estrogen regulation of epidermal growth factor receptor messenger ribonucleic acid. Mol Endocrinol 1988;2:230-5. 20. Murphy LJ, Murphy LC, Friesen HG. Estrogen induces insulin-like growth factor-I expression in the rat uterus. Mol Endocrinol 1987;1:445-50. Y, Riemer RK, Roberts JM. The concentration of 21. Sadovsky estrogen receptors in rabbit uterine myocytes decreases in culture. AM J OBSTET GYNECOL 1992;167:1631-5. 22. Lumsden MA, West CP, Bramley T, Rumgay L, Baird DT. The binding of epidermal growth factor to the human uterus and leiomyomata in women rendered hypoestrogenie by continuous administration of an LHRH agonist. Br J Obstet Gynaecol 1988;95:1299-304. 23. Rein MS. Friedman Al. Pandian MR. Heffner LT. The secretion of insulin-like growth factors I and II by explant culture of fibroids and myometrium from women treated with gonadotropin-releasing hormone agonist leuprolide acetate. Fertil Steril 1990:46:1554-8. LC, Irwin JC, Dsupin BA, et al. Insulin-like growth 24. Giudice factor (IGF), IGF binding protein (IGFBP), and IGF receptor gene expression and IGFBP synthesis in human uterine leiomvomata. Hum Reorod 19938: 1796-806. 25. Hu J, Surti U: Subgroups of uterine leiomyomas based on cytogenetic analysis. Hum Path01 1991;22: 1009-16. .,I

0