Hyperinsulinemia

and Human Chorionic Gonadotropin Synergistically the Growth of Ovarian Follicular Cysts in Rats

Promote

Leonid Poretsky, Jeffrey Clemons, and Katryna Bogovich Tonically elevated serum luteinizing hormone (LH) and hyperinsulinemia are prominent features of polycystic ovary syndrome (PCO) in women, but the relative roles of LH and insulin in the pathogenesis of PC0 is still unknown. The present study was undertaken to determine the effect(s) hyperinsulinemia might have on the induction of follicular cysts by LH/ human chorionic gonadotropin (hCG) in the rat. Beginning on day 65 of age, adult female rats were given one of the following in vivo treatments: (1) vehicle alone; (2) a high-fat diet to control for the effects of weight-gain; (3) up to 6 U insulin per day; (4) 50 Pg gonadotropin-releasing hormone (GnRH) antagonist (GnRHant) per day; (5) 1.5 IU hCG twice daily; (6) insulin + hCG; (7) insulin + GnRHant; (6) hCG + GnRHant; or (9) hCG + insulin + GnRHant. After 22 days of treatment, animals were killed on day 23, trunk blood was collected, and ovaries were excised for histological study. Regular cycles, assessed by vaginal smears, ceased after 10 days for most animals in treatment groups receiving hCG, but continued in all other treatment groups. All the animals in each hCG-treated group developed either unilateral or bilateral cystic ovaries, while no animals in the groups not receiving hCG developed follicular cysts. More animals from each group treated with both hCG and insulin possessed bilateral ovarian cysts than did rats treated with hCG alone: 60% and 60%. respectively. Mean ovarian product ([MOP] OP = ovarian length x width, mm*) was greater for animals that received hCG + insulin than for animals that received hCG alone (46.6 2 5.1 and 32.5 + 2.1, respectively; P c .Ol). Animals treated with hCG + GnRHant f insulin displayed more follicular cysts than animals treated with hCG f insulin (7.9 f 1.3 and 4.5 f 1.27 cysts/rat, respectively; P < .05). In addition, only rats that received hCG + insulin + GnRHant possessed a greater mean follicular cyst volume than rats that received hCG alone (2.16 & 0.56 and 1.0 + .19 mm3, respectively; P < .05). Rats treated with insulin had elevated plasma insulin concentrations (range, 100 to 340 plU/mL), while all other treatment groups had plasma insulin values similar to those of control rats (-50 pIU/mL). Plasma estradiol concentrations were elevated only for animals treated with hCG alone (201 + 17 pg/mL compared with 144 f 30 pg/mL for all other treatment groups, P < .05). Plasma estrone concentrations were elevated only for rats treated with insulin (139 2 11 compared with 95 f 9 pg/mL for all other treatment groups; P < .02). Androstenedione concentrations were elevated for rats that received hCG when compared with values for other treatment groups (13.4 f 1.0 and 10.7 f 0.4 ng/mL, respectively; P < .05). However, treatment with both hCG and insulin failed to elevate plasma androstenedione values further (14.6 + 1.1 ng/mL). These data indicate that while unabated LH-like stimulation is sufficient to induce ovarian follicular cysts in intact rats, hyperinsulinemia enhances follicular cyst development in this species by increasing the volume of the induced cysts and thus increasing ovarian size. Copyright Q 1992 by W.B. Saunders Company

P

OLYCYSTIC OVARY syndrome (PCO) is a common disorder characterized by chronic anovulation, bilateral polycystic ovaries, “tonic” luteinizing hormone (LH) secretion, and diverse degrees of hyperandrogenism.1-5 Although much research has been performed in this area, the cause of PC0 remains unknown. This has been due, in part, to the lack of a satisfactory animal model for this disease. In addition to women, ovarian cysts occur naturally in many other mammals such as cows, cats, dogs, and rats.6-i0 These observations have led to the use of rodents to unravel the mechanisms involved in the onset of the cystic ovary state.“.‘” However, none of the PC0 models described to date have taken into consideration the observation that most patients with PC0 are insulin resistant and thus hyperinsulinemic.2i-24 This observation has led to the hypothesis that hyperinsulinemia is responsible, at least in part, for the morphological and biochemical characteristics of ovaStill, the precise role of ries from women with PCO. 25~26 hyperinsulinemia in the induction of cystic ovaries is yet to be determined. Unabated stimulation by low doses of LH-like activity (human chorionic gonadotropin, hCG) induces ovarian follicular cysts in both intact immature and pregnant rats, but noi in hypophysectomized rats.i3-15 In contrast, experimental hyperinsulinemia results in down-regulation of rat ovarian insulin receptors and up-regulation of type I insulin-like growth factor (IGF-I) receptors, but fails to Mefabohsm, Vol 41, No 8 (August), 1992: pp 903-910

induce polycystic ovaries.” Therefore, hyperinsulinemia does not appear to play an obligatory role in the induction of the ovarian cystic state in rats. Still, hyperinsulinemia may enhance the ability of LHihCG to induce polycystic ovaries. The following study was performed to test this hypothesis. In this study, a gonadotropin-releasing hormone (GnRH) antagonist (Nal-Glu) was used to suppress both estrogeninduced and basal LH secretion. As with the progesteronesynchronized immature rat*” and the pregnant rat,l4 this model allows one to observe the effects of unabated LH and insulin stimulation on ovarian morphology without the

From the Divisions of Endocrinology, Departments of Medicine, Beth israel Medical Center and Mount Sinai School of Medicine, New York, NY,: Cab&i Medical Center, New York, NY,: New York Medical College, Valhalla, NY: and the Depatiment of Obstettics and Gynecology. University of South Carolina School of Medicine, Columbia, SC. Supporled in pati by Herbert H. Singer Collaborative Research Award at Beth Israel Medical Center (L.P.); the Roberro E. Pope Fund ut Cabrini Medical Center; and National Institutes of Health Grant No. HD-22738 (L.P.). Presented in part at the 73rd Annual Meeting of the Endocrine Society, Washington, DC, June 19-21. 1991. Address reprint requests to Leonid Poretsky MD, Division of Endocrinology, Cabrini Medical Center, 227 19th St, New York, NY 10003. Copyright 0 I992 by W.B. Saunders Cornpan?, 0026049519214108-0017$03.00l0 903

PORETSKY, CLEMONS, AND BOGOVICH

904

Table 2.

neuroendocrine feedback problems encountered in other models for cyst development.11~12J6-20~28~z9 Furthermore, comparison of the results of the present study with that of the progesterone-synchronized rat permitted an evaluation of the effects of elevated, but physiological, concentrations of serum progesterone on the induction of ovarian follicular cysts. MATERIALS

Dw

AND METHODS

Experimental Animals The following in vivo procedures were performed in the Beth Israel Medical Center Animal Facility according to a protocol approved by the Committee for Scientific Activities of Beth Israel Medical Center. A total of 90 female Sprague-Dawley rats (Taconic Laboratories, New York. NY) were housed at 54 2 3 days of age in cages in rooms with 12-hour light/dark cycles. Initially, all animals were given rat chow and water ad libitum. For 31 days, the animals underwent daily vaginal smears to document consecutive 4- to 5-day estrous cycles. On day 85 2 3 days of age, the rats were randomly distributed into nine in vivo treatment groups as shown in Table 1. Briefly, vehicle (phosphate-buffered saline [PBS] containing 0.09% pig skin gelatin, pH 7.0) was administered subcutaneously twice daily throughout the in vivo treatment period. Insulin (NPH-Humulin, U-100; Lilly, Indianapolis, IN) was administered as previously described to obtain a hyperinsulinemic state without hypoglycemia. 27As shown in Table 2, 0.5 U insulin was administered on the first day of in vivo treatment, and this dose was gradually increased during the 22-day treatment period until 6 U insulin were administered per day. “Nal-Glu,” the GnRHantagonist (GnRHant) was generously provided by Dr Jean Rivier (The Salk Institute, San Diego, CA), and 50 t.rgwas administered once daily throughout the treatment period. Preliminary studies by Dr Rivier (personal communication, October 1989) indicated that this dose of GnRHant is sufficient for full suppression of LH secretion in rats. hCG (Sigma Chemical, St Louis, MO) was administered twice daily to provide a total dose of 3.0 IUld. This dose of hCG has induced ovarian follicular cysts within 9 days in progesterone-synchronized immature rats.t3 Daily vaginal smears were obtained to evaluate individual cyclicity. In addition, individual rat weights were measured daily.

Schedule of Insulin injections

0.5

2

1.5

3

1.5

4

2.0

5

2.5

6

3.0

7

4.0

a

4.0

9

5.0

10

5.0

11-22

6.0

NOTE. NPH-Humulin,

a day

Insulin Dose

1

(U/d)

U-100 was administered subcutaneously twice

in equal doses.

Animals were decapitated on the 23rd day of treatment, approximately 12 hours after the last injection. Trunk blood was collected in heparinized polypropylene tubes for measurement of plasma hormone levels. To obtain adequate amounts of plasma for insulin, hCG. and steroid assays, each group of 10 animals was divided into five pairs and equal aliquots of plasma from the two animals in each pair were pooled. Ovaries were removed, their dimensions were measured, and both ovaries from each animal were fixed in formalin. Sections were prepared and stained with hematoxylineosin by the Histology Core Facility at the University of South Carolina School of Medicine. Photomicrographs were obtained with a 2.5x objective on a Zeiss 351M microscope (Carl Zeiss, Oberkochen, Germany). Radioimmunoassays Before analysis, blood samples were centrifuged at 1,000 x g and the resulting supernatants were transferred to individual polypropylene tubes. Plasma insulin, hCG, and free testosterone were measured using radioimmunoassay (RIA) kits obtained from Diagnostic Products (Los Angeles, CA). Sex hormone-binding globulin (SHBG) was measured using an RIA kit obtained from Diagnostic Systems Laboratories (Webster, TX). To determine total plasma steroid concentrations. known volumes of gel-PBS and plasma (each containing 1,000 cpm of 3H-estrone [New England Nuclear, Boston, MA] to monitor

Table 1. Treatment Groups, Diets, and Mean Weight Changes In Viva Treatment

Diet

Initial Weight

Final Weight

Vehicle alone*

Group

RCSD

232.5 + 3.8

250.2 + 7.6

17.7 2 5.2

High fat

HFD-SD

232.1 it 4.6

253.4 + 6.2

21.3 + 4.0

lnsulint

RCSD

232.5 2 7.1

270.3 + 8.1

37.8 + 2.4

GnRHant*

RCSD

232.3 2 5.6

262.0 + 4.8

29.7 2 2.9

Weight Change

Insulin + GnRHant

RCSD

232.4 k 5.9

272.2 + 7.1

39.8 k 4.4

hCG§

RCSD

232.1 + 2.0

256.2 + 5.2

24.1 + 4.6 20.7 f 2.6

hCG + GnRHant

RCSD

231.9 + 5.3

252.6 _f 4.6

hCG + insulin

RCSD

232.1 ? 6.0

264.5 2 6.1

32.4 ? 3.2

hCG + insulin + GnRHant

RCSD

232.0 k 5.6

264.1 + 6.6

32.1 +- 2.3

NOTE. Values are grams + SEM; n = 10 for all in viva treatment groups. Abbreviations: RCSD, rat chow, sugar cubes, 5% dextrose; HFD-SD, high-fat diet, sugar cubes, 5% dextrose (HFD is a paste with 50% of its calories as fat; ICN Biomedical, Costa Mesa, CA). *All injections were administered subcutaneously in 0.2 mL. tSee Table 2 for insulin dosage regimen. $50 Fg by injection once daily. 91.5 IU by injection twice daily (3 IlJld total).

HYPERINSULINEMIA

AND hCG SYNERGISM

905

IN PC0

procedural recoveries) were extracted individually with diethyl ether. The ether phase of the extractions was evaporated to dryness and the residues were reconstituted with 1 mL gel-PBS. Procedural recoveries were assessed by counting 0.2 mL of each of the reconstituted samples in a Beckman LSl50 scintillation counter (Beckman Instruments, Norcross, GA). The remaining portion of each sample was analyzed by RIA for progesterone, androstenedione. estradiol, and estrone content. Antisera for estradiol was generously provided by Dr Gordon Niswender (Fort Collins, CO). Antisera against the bovine serum albumin (BSA) conjugate of 1la-(p-hydroqphenyl) acetoxyandrostenedione and the hemisuccinate derivative of this steroid were the generous gifts of Dr Gerry Nordblom and Dr Barry England (Ann Arbor, MI). Antisera for estrone and the iodinated estrone ligand were obtained from Pantex (Santa Monica, CA). RIAs for androstenedione and estradiol were performed using established procedures.‘3,30.31The estrone RIA was performed using the protocols described for the androstenedione assay.31

Stufistics Ovarian product (OP) was calculated by multiplying the largest dimensions for ovarian length and width, yielding units of mm2. Mean ovarian product (MOP) was calculated as the mean value of OPs for each group of animals (n = 20 ovaries). The differences between mean values of rat weights, between mean values of ovarian products, and between mean values of plasma hormone considered to concentrations were evaluated by t test, with P < .05 indicate a statistical difference between groups. RESULTS

Animal Weights Table 1 shows the effect of the various in vivo treatments on mean animal treatment group weights. All animals gained weight during the in vivo treatment period. The largest weight-gains were observed for rats in groups that received insulin, and these changes in weight were significantly greater than the weight changes of the other in vivo treatment groups (P < .025). There were no statistical differences in the observed weight-gains among the groups

Control Fig 1. Gross morphology GnAHant.

hCG of representative

that received insulin or among the groups that did not receive insulin. Ovarian Morphology In this series of experiments, the term cystic refers to follicles generally greater than 0.8 mm in diameter, with well-developed thecal shells and just a remnant of granuloss cells. Furthermore, ovaries with cystic follicles also possessed interstitial tissue with a stimulated appearance. Figure 1 illustrates the effects of the in vivo treatments on gross ovarian morphology. Ovaries from animals treated with hCG alone possessed cystic follicles and appeared significantly larger than ovaries from control rats. Ovaries from rats treated only with insulin possessed no cystic follicles and were approximately the same size as ovaries from control rats. In contrast, ovaries from rats treated with both hCG and insulin with (shown) or without (not shown) GnRHant were significantly larger than ovaries from all other treatment groups and displayed very large cystic follicles. Histological examination of representative ovaries from all nine treatment groups show the striking difference between ovaries from animals that did not receive hCG (Fig 2) and those that did (Fig 3). These figures clearly demonstrate that only animals from groups that received hCG developed ovarian follicular cysts by the end of the in vivo treatment period (Fig 3). While all animals in all hCGtreated groups possessed cystic ovaries, 40% of the animals treated with hCG alone developed unilateral cystic ovaries, while only 10% to 20% of animals treated with hCG plus either insulin, GnRHant, or insulin + GnRHant developed cysts unilaterally. Overall, animals with unilateral ovarian cysts usually displayed much fewer cysts than animals with bilateral cysts (2.1 2 0.86 and 7.39 * 0.68 cysts/rat, respectively; P < .05). In addition, comparison of animals with only bilateral cystic ovaries indicated that rats treated with hCG alone possessed fewer cysts than rats treated with

Insulin

hCG + Insulin + GnRHant

ovaries from rats treated with vehicle alone; hCG alone; insulin alone; and hCG + insulin f

PORETSKY, CLEMONS, AND BOGOVICH

;

Fig 2. Histological sections of representative ovaries on day 23 of treatment from rats not treated with hCG. (A) vehicle alone; (6) high-fat diet; (C) insulin; (D) GnRHant; and (E) insulin + GnRHant. Photomicrographs were taken with a 2.5x objective.

either hCG + GnRHant or hCG + insulin + GnRHant (4.2 + 1.1, 8.3 2 0.8. and 9.6 ? 1.3 cysts per rat, respectively; P < .05). The number of cysts in the ovaries of rats treated with hCG + insulin was highly variable (6.5 -+ 1.7) and was not significantly different from the number of cysts observed in rats treated with hCG alone. These relationships did not change when ovaries from unilaterally cystic animals were included in the analyses. As shown in Table 3, the mean volume of folhcular cysts from animals treated with hCG + insulin + GnRHant was significantly greater than the mean volume of cysts from animals treated with hCG alone (2.2 k 0.6 and 1.0 + 0.2 mm3, respectively; P < .05). Mean volumes of cysts from rats treated with hCG + insulin or hCG + GnRHant were highly variable and were not significantly different from

Fig 3. Histological sections of representative ovaries on day 23 of treatment from rats treated with hCG. (A) hCG; (6) hCG + GnRHant; (C) hCG + insulin; (0) hCG + insulin + GnRHant. Photomicrographs were taken with a 2.5x objective.

values observed for cysts from animals treated with hCG alone. In contrast, not only were the MOPS for all hCGtreated groups greater than the MOPS for the groups that did not receive hCG (P < ,025) but the MOPS for the groups treated with hCG + insulin with or without GnRHant were significantly greater than the MOPS for all other groups (P < .025). Reproductive Cycles As shown in Table 4, reproductive cycles assessed by vaginal smears remained regular in all rats who did not receive hCG or GnRHant. Only 10% to 30% of rats who received hCG. GnRHant, or a combination of hCG and GnRHant ? insulin displayed vaginal smears indicative of normal cycles during the in vivo treatment period.

HYPERINSULINEMIA

AND hCG SYNERGISM IN PC0

907

Table 3. Ovarian Morphology No. of RatsWith CysticOvaries Bilateral

Unilateral

Mean CystVolume Immx)2 SEM

Vehicle

0

0

NA

23.3 2 1.3

High fat

0

0

NA

23.7 + 1.3

Insulin

0

0

NA

24.4 ? 1.6

GnRHant

0

0

NA

15.0 f 0.8

Insulin + GnRHant

0

0

NA

14.8 + 0.9

hCG

6

4

1.0 + 0.2

32.5 + 2.1t

hCG + GnRHant

9

1

1.8 2 0.9

36.1 r 2.8t

hCG + insulin

8

2

2.0 -t 0.9

46.6 + 5.1$

hCG + insulin + GnRHant

8

2

2.2 + 0.6*

44.7 + 6.5$

TreatmentGroup

MOP (mm?) 5 SEM

NOTE. n = 10 rats or 20 ovaries for all in vivo treatment groups. Mean cyst volume for each group was calculated by determining the mean cyst diameter for each rat and then calculating the mean cyst volume for the animal (4.19r3). The means of these volumes were then obtained for each treatment group. MOP is the product of the two largest ovarian axes. *Significantly different from all groups not treated with hCG and from the group treated with hCG alone, P < .025. tsignificantly different from all groups not treated with hCG, P < ,025. *Significantly different from all groups not treated with hCG + insulin, P < .025.

Serum Hormone Concentrations

As shown in Fig 4A, plasma insulin concentrations ranged from 152.6 ? 20.8 to 315.7 ? 61.2 uU/mL in groups that received insulin injections. These values were significantly greater than those observed for rats in groups that did not receive exogenous insulin ( I 55 uU/mL, P < .025). Similarly, plasma hCG concentrations were detectable only in animals that received hCG injections (Fig 4B). These values ranged between 26.2 ? 12.2 and 108.0 c 63.6 mIU/mL and were significantly greater than values for groups that did not receive hCG (P < .05). Plasma estradiol concentrations (Fig 4C) were significantly elevated for rats that received either hCG alone (201 l?r.17 pg/mL), hCG + GnRHant (235 +- 16 pg/mL), or hCG t GnRHant + insulin (207 + 27 pg/mL) compared with all other treatment groups (averaging - 143 + 6 pg/ mL, P < .05 for all comparisons). In contrast, plasma estrone concentrations (Fig 4D) were elevated significantly above control values only for rats treated only with insulin (139 i 11 and 95 i 9 pg/mL, respectively; P < .02). Androstenedione plasma concentrations (Fig 4E) were elevated above control values for all groups that received hCG (averaging - 14.19 * 0.66 \-a 10.74 t 0.39 ng/mL, respectively; P < .05). Androstenedione values did not increase further when rats were treated with both hCG and insulin. Table 4. Reproductive Cycles In Viva Treatment Group Vehicle alone

% Rats MaintainingCycles 100

High-fat diet

100

Insulin

100

GnRHant

30

Insulin + GnRHant

20

hCG

10

hCG t GnRHant

20

hCG + insulin

30

hCG t insulin + GnRHant

30

NOTE. See text for significant differences.

Free testosterone and SHBG plasma concentrations were virtually undetectable for all treatment groups (data not shown).

DISCUSSION

The role of insulin resistance and consequent hyperinsulinemia in the pathogenesis of PC0 remains unknown. High doses of insulin stimulate ovarian steroidogenesis both in vitro and in vivoz5J6 and potentiate the stimulatory effects of LH on theta-interstitial cells in vitro.32*3’These observations and the observation that many women with PC0 display hyperinsulinemia have led to the hypothesis that hyperinsulinemia may be an important factor in the development of ovarian follicular cysts.Z4,26However, the precise role that hyperinsulinemia may play in the development or the perpetuation of the PC0 state is yet to be determined. Previous studies have shown that experimentally induced hyperinsulinemia is not sufficient to induce ovarian follicular cysts in the rat, even though such treatment did lead to down-regulation of ovarian insulin receptors and upregulation of ovarian IGF-I receptorsz7 In contrast, unabated stimulation by low doses of LH-like activity in intact progesterone-synchronized immature rats and pregnant rats does lead to the formation of follicular cysts.13J4 Together, these observations suggest that unabated stimulation by LH rather than hyperinsulinemia plays the major role in the induction of cystic ovaries in the rat. The present series of experiments were designed to determine if hyperinsulinemia potentiated the effect of unabated LH-like stimulation on ovarian follicular cyst development in the rat. GnRHant was used in this series of experiments to avoid an estradiol-induced LH surge during the initial hCG treatments. The effects of unabated hCG stimulation on ovarian histology under these conditions are similar to those observed in the progesterone-synchronized immature rat and the pregnant rat,‘sJ4 and support the concept that the elevated serum progesterone concentra-

908

PORETSKY, CLEMONS, AND BOGOVICH

ABCDEFGHI

^ :

250 200

: ;

150

n :

100

5

50

c ABCDEFGHI

ABCDEFGHI

AECDEFGHI Treatment

Groups

Fig 4. Plasma hormone concentrations on day 23 of treatment. X-axis labels: (A) vehicle alone; (B) high-fat diet; (C) insulin; (D) GnRHant; (E) insulin + GnRHant; (F) hCG; (G) hCG + GnRHant; (H) hCG + insulin; (I) hCG + insulin + GnRHant. Plasma hormone values were obtained by RIA as described in Methods. *Significantly different from vehicle alone (P < .BB).

tions achieved in previous studies do act primarily to suppress LH secretion and not to suppress follicular development directly at the level of the ovary.13J4.34 Some animals did tend to escape from the regulation of the GnRHant during the course of these in vivo treatments, as evidenced by vaginal smears and the presence of corpora lutea in these animals. However, this effect was most probably due to the estradiol feedback effects simply being greater than the effects of the dose of antagonist used in these animals. The ability of unabated hCG stimulation to induce cystic follicles in the absence of GnRHant supports similar observations in a previous study.35 The hCG-only treated rats tended to exhibit more unilateral cystic ovaries and smaller follicular cysts than the groups treated with hCG plus either GnRHant, insulin, or a combination of the three. This is not explained by a tendency for rats treated only with hCG to escape the cystic state, since only one of the 10 animals in this group provided vaginal smears indicative of a maintained cycling ability. Rather, the data seem to suggest that the GnRHant and progesteronel3J4 may be regulating factor(s) that influence the development of large rather than small cystic follicles. Inclusion of insulin with the in vivo hCG + GnRHant treatments resulted in both the development of the largest ovarian cysts and the largest ovaries for any treatment group. At present, we can only speculate on the mecha-

nisms by which insulin exerts its effects on cyst development in these animals. The effects of insulin on ovarian and ovarian cyst size may be due, in part, to the ability of insulin to up-regulate ovarian LH/hCG receptors.25s26.32.33 In addition, ovarian growth may be promoted by insulin due to a potential increase in ovarian IGF-I receptors, as previously described for hyperinsulinemic rats.27 Although insulin is most likely to act through its own receptor under normal circumstances,26,36,37sufficiently elevated concentrations of insulin may promote ovarian growth by binding to and acting through the ovarian IGF-I receptors.38-42 If this is the case, one might speculate that growth hormone or IGF-I may also act synergistically with hCG to promote the growth of ovarian cysts. Our observations regarding the effect of insulin on ovarian size are supported by a recent report of a 16-year-old with type B syndrome of insulin resistance and acanthosis nigricans.43 Sonography of the ovaries of this patient showed a greater than twofold increase in ovarian volume during 35 days of high-dose insulin therapy. This observation, together with the results obtained in our present study, indicates that hyperinsulinemia may have a profound effect on ovarian size. As expected, unabated stimulation by hCG increased several plasma steroid concentrations, including estradiol and androstenedione. In addition, insulin by itself increased plasma estrone concentrations-an effect probably due to stimulation of aromatase activity.27 The inability of

HYPERINSULINEMIA

909

AND hCG SYNERGISM IN PC0

hyperinsulinemia alone or in the presence of unabated hCG stimulation to further increase plasma androstenedione concentrations is very interesting, given the data on the effects of insulin on ovarian androgen production in vitro.32933 However, since both the duration and magnitude of hyperinsulinemia achieved in this series of experiments were limited, the present results do not exclude the possibility that a prolonged exposure to an “extreme” hyperinsulinemic environment (such as insulin values in the range > 1,000 mIU/mL-as in patients with syndromes of insulin resistance and acanthosis nigricans26 may induce, by itself, polycystic ovaries by interfering with normal neuroendocrine, ovarian, or both neuroendocrine and ovarian functions. In summary, this series of experiments indicate that hyperinsulinemia affects the development of cystic ovaries in the rat by causing an increase in the size of the cystic follicles and in the size of the ovary. These effects of insulin may be due to its properties as a growth factor. However, in

addition, both hCG and insulin treatments did lead to an increase in plasma estrogen concentrations (indicating an increase in ovarian aromatase activity). Finally, the animal model described in this study simulates several important features of PCO: (1) unabated stimulation by LH-like activity; (2) hyperinsulinemia; (3) large cystic ovaries; (4) anovulation; and (5) elevated serum androstenedione. Therefore, this model may be appropriate for further studies regarding the role of insulin in the etiology and pathogenesis of this common reproductive disorder. ACKNOWLEDGMENT

The authors are indebted to Dr Jean Rivier of The Salk Institute for the generous gift of GnRHant; to Dr Andrea Dunaif of the Mount Sinai School of Medicine for her advice regarding the use of GnRHant; to the personnel of the Animal Facility at Beth Israel Medical Center; to the Histology Core Facility at the University of South Carolina School of Medicine; and to Olga Rosa-Desrosiers for her expert secretarial assistance.

REFERENCES

1. Yen SSC: The polycystic ovary syndrome. Clin Endocrinol 12:177-208, 1980 2. Coney P-J: Polycystic ovarian disease: Current concepts of pathology and therapy. Fertil Steril42:667-682.1984 3. Futterweit W: Polycystic ovarian disease, in Buchsbaum HJ (ed): Clinical Perspectives in Obstetrics and Gynecology Series. New York, NY, Springer-Verlag, 1984 4. McKenna TJ: Pathogenesis and treatment of polycystic ovary syndrome. N Engl J Med 318558-562, 1988 5. Yen SSC: Polycystic ovary syndrome, in Yen SCC, Jaffe RB (eds): Reproductive Endocrinology. New York, NY, Saunders, 1986 6. Roberts SJ: Clinical observations on cystic ovaries in dairy cattle. Cornell Vet 45:497-513,1955 7. Stein BS: Genital system, in Catcott EJ (ed): Feline Medicine and Surgery. Santa Barbara, CA, American Veterinary Publications, 1975 8. Peluso JJ, Steger RW, Huang H, et al: Pattern of follicular growth and steroidogenesis in the ovary of aging cycling rats. Exp Aging Res 5:319-333,1979 9. Peluso JJ, England-Charlesworth C: Formation of ovarian cysts in aged irregularly cycling rats. Biol Reprod 24:1183-1190. 1981 10. Kesler DJ, Garverick HA: Ovarian cysts in dairy cattle: A review. J Anim Sci 5.5:1147-1159, 1982 11. Mahesh VB, Mills TM, Bagnell CA, et al: Animal models for study of polycystic ovaries and ovarian atresia, in Mahesh VB, Dhindsa DS, Anderson E, et al (eds): Regulation of Ovarian and Testicular Function. New York, NY, Plenum, 1987, pp 237-258 12. Mahajan DK: Polycystic ovarian disease: Animal model. Endocrinol Metab Clin Am 17:705-732,1988 13. Bogovich K: Induction of follicular cysts in progesterone synchronized immature rats: Evidence that suppression of follicular aromatase activity is not a prerequisite for the induction of cystic follicles. Endocrinology 124:1646-1653, 1989 14. Bogovich K: Induction of ovarian follicular cysts in the pregnant rat by human chorionic gonadotropin. Biol Reprod 45:34-42,199l 15. Bogovich K: Luteinizing hormone (LH)-like activity and follicle-stimulating hormone (FSH) are sufficient for the induction

of cystic ovaries in hypophysectomized rats, in Hirshfield AN (ed): Growth Factors and the Ovary. New York, NY, Plenum, 1989, pp 345-349 16. Leathem JH: Hormonal influences on the gonadotropin sensitive hypothyroid rat ovary. Anat Ret 131:487-500, 1958 17. Swanson H, van der Werff ten Bosch JJ: The “early androgen” syndrome, its development and the response to hemispaying. Acta Endocrinol44:1-12,1964 18. Barraclough CS: Modification in the CNS regulation of reproduction after exposure of prepubertal rats to steroid hormones. Recent Prog Horm Res 22:503-515,1966 19. Adams WC, Senseman DM: Aromatization of androstenedione by induced polycystic ovaries in the rat. J Steroid Biochem 7:309-310,1976 20. Brawer JR, Naftolin F, Martin J, et al: Effects of a single injection of estradiol valerate on the hypothalamic arcuate nucleus and on reproductive function in the female rat. Endocrinology 103:501-502.1978 21. Chang RJ, Nakamura RM, Judd HL, et al: Insulin resistance in nonobese patients with polycystic ovarian disease. J Clin Endocrinol Metab 57:356-359, 1983 22. Dunaif A, Segal KR, Futterweit W, et al: Profound peripheral insulin resistance, independent of obesity. in polycystic ovary syndrome. Diabetes 38:1165-1174,1989 23. Burghen GA, Givens JR, Kitabchi AE: Correlation of hyperandrogenism with hyperinsulinism in polycystic ovarian disease. J Clin Endocrinol Metab 50:113-116,198O 24. Barbieri RL, Smith S, Ryan KJ: The role of hyperinsulinemia in the pathogenesis of ovarian hyperandrogenism. Fertil Steril 50:197-212. 1988 25. Poretsky L, Kalin M: The gonadotropic function of insulin. Endocr Rev 8:131-141,1987 26. Poretsky L: On the paradox of insulin-induced hyperandrogenism in insulin resistant states. Endocr Rev 12:3-13, 1991 27. Poretsky L, Glover B, Laumas V, et al: The effects of experimental hyperinsulinemia on steroid secretion, ovarian 125-1insulin binding and ovarian 125-I-Insulin-like growth factor I binding in the rat. Endocrinology 122:581-585, 1988 28. Copman TL, Adams WC: Relationship of polycystic ovary

PORETSKY, CLEMONS, AND BOGOVICH

induction to prolactin secretion: Prevention of cyst formation by bromocriptine in the rat. Endocrinology 108:1095-1097,198l 29. Lee M-T, Bruot BC, Adams WC: Hormonal changes in the early development of ovarian cysts in the rat. Biol Reprod 35542-548, 1986 30. Nordblom GD, Webb R, Counsel] RE. et al: A chemical approach to solving bridging phenomenon in steroid radioimmunoassays. Steroids 38:161-173, 1981 31. Bogovich K, Scales LM, Higginbottom E, et al: Short term androgen production by rat ovarian follicles and long term steroidogenesis by thecal explants in culture. Endocrinology 118:1379-1386, 1986 32. Cara JF, Rosenfield RL: Insulin-like growth factor I and insulin potentiate luteinizing hormone-induced androgen synthesis by rat ovarian thecal-interstitial cells. Endocrinology 123:733739.1988 33. Hernandez ER, Resnick CE, Holtzclaw DW, et al: Insulin as a regulator of androgen biosynthesis by cultured rat ovarian cells: Cellular mechanism(s) underlying physiological and pharmacological hormonal actions. Endocrinology 127:2034-2043, 1988 34. Bogovich K, Richards JS, Reichert LE Jr: Obligatory role of luteinizing hormone (LH) in the initiation of preovulatory follicular growth in the pregnant rat: Specific effects of human chorionic gonadotropin and follicle-stimulating hormone on LH receptors and steroidogenesis in theta, granulosa, and luteal cells. Endocrinology 109:860-867, 1981 35. Bogovich K: Can an ovarian cystic state induced by human

chorionic gonadotropin (hCG) cause abnormal changes in rat endometrial morphology? 72nd Annual Meeting of the Endocrine Society, Atlanta, GA, June 20-23, 1990 (abstr no. 170) 36. Poretsky L, Smith D, Seibel M. et al: Specific insulin binding sites in human ovary. J Clin Endocrinol Metab 59:809-811. 1984 37. Poretsky L, Grigorescu F, Moses AC, et al: Distribution and characterization of the insulin and IGF-1 receptors in the human ovary. J Clin Endocrinol Metab 61:728-734, 1985 38. Fradkin JE, Eastman RC, Lesniak spillover at the hormone receptor-Exploring disease. N Engl J Med 320:640-645,1989

MA. et al: Specificity its role in human

39. Poretsky L: Specificity-spillover phenomenon ovary. N Engl J Med 321:474. 1989 (letter)

in the human

40. Gates GS. Bayer S, Seibel M, et al: Characterization insulin-like growth factor I binding to human granulosa obtained during in vitro fertilization. J Recept Res 7:885-902,

of cells 1987

41. Poretsky L, Bhargava G, Levitan E: Type I insulin-like growth factor receptors in human ovarian stroma. Horm Res 33:22-26, 1990 42. Pepper GM, Poretsky L, Gabrilove JL. et al: Ketoconazole reverses hyperandrogenism in a patient with insulin resistance and acanthosis nigricans. J Clin Endocrinol Metab 65:1047-1052, 1987 43. DeClue TJ, Shah SC, Marchese M, et al: Insulin resistance and hyperinsulinemia induce hyperandrogenism in a young type B insulin-resistant female. J Clin Endocrinol Metab 72:1308-1311, 1991

Hyperinsulinemia and human chorionic gonadotropin synergistically promote the growth of ovarian follicular cysts in rats.

Tonically elevated serum luteinizing hormone (LH) and hyperinsulinemia are prominent features of polycystic ovary syndrome (PCO) in women, but the rel...
2MB Sizes 0 Downloads 0 Views