High frequency of IgG antagonizing follicle-stimulating hormone-stimulated steroidogenesis in infertile women with a good response to exogenous gonadotropins

High frequency of IgG antagonizing follicle-stimulating hormone-stimulated steroidogenesis in infertile women with a good response to exogenous gonadotropins

FERTILITY AND [email protected] VOL. 69, NO. I, JANUARY 1998 Copyright 01998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Pr...

829KB Sizes 4 Downloads 26 Views

FERTILITY AND [email protected] VOL. 69, NO. I, JANUARY 1998 Copyright 01998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.

High frequency of IgG antagonizing follicle-stimulating hormonestimulated steroidogenesis in infertile women with a good response to exogenous gonadotropins Yves Reznik, M.D.,* Annie Benhafm, Ph.D.,7 Rbmy Morel/o, M. D.,$ Michel Hetiicoviez, M.D.,5 Jean-Jacques Ballet, Ph.D. M.D.,1)and Jacques Mahoudeau M.D.*

Received January 28, 1997; revised and accepted August 12, 1997. Supported by a grant from the French Minister for Research and Education in the “Life Sciences and Health” Program, grant number 93-837. Presented at the International symposium on life cycle of the ovarian follicle, Fort Lauderdale, Florida, November 2-5, 1995, and at the X International Congress of Endocrinology, San Francisco, California, June 12-151996. Reprint requests: Yves Re.&ik, M.D., Service d’Endocrinoloaie. CHU C&e de Nacr< t4033 Caen Cedex, France. * Service d’Endocrinologie, CHU CBte de Nacre. 14033 t Laboratoire de Biochimie Ep 9 CNRS CHU C&e de Nacre, 14033 Caen. $ Laboratoire d’lnformatique Medicale et d’Epidemiologie, CHRU Ciemenceau, 14033 Caen. jj Service d’ksistance Medicale a la Procreation, CHRU Clemenceau, 14033 Caen.

)/Laboratoire d’lmmunologie, CHRU Clemenceau, 14033 Caen. 001%0282/98/519.00 PII 5001s0282(97)00430-5

46

*Service d’Endocrinologie, tLaboratoire de Biochimie Ep 9 CNRS, CHU C&e de Nacre, Caen, and @en/ice d’Assistance Mkdicale B la Proctiation, SLaboratoire d’informatique M6dicale et d’Epid6miologie and [[Laboratoire d’lmmunologie, CHRU Clkmenceau, Caen, France

Objective: To investigate the presence of FSH-blocking IgG in infertile women. Design: Retrospective study. Sera from patients and controls were processed for IgG purification, and purified IgG were tested at various concentrations for their ability to inhibit the recombinant human FSH-induced P production in vitro by human granulosa cells. Setting: Departments of Endocrinology, and Obstetrics and Gynecology, University of Caen. Patient(s): Fifty-seven infertile women including 14 women with premature ovarian failure (POF), 29 women with a poor response to IVF-ET, and 14 women with a good response to IVF-ET. Controls consisted of 22 healthy age-matched women. Intervention(s): IVF-ET allowed human granulosa cell pooling and culture for FSH bioassay. Main Outcome Measure(s): Inhibition by purified IgG of the in vitro recombinant human FSH-induced P production by human granulosa cells. Result(s): Blocking IgG were identified in only 3 of 14 POF and in 2 of 29 women with a poor response to IVF-ET. In contrast, IgG from women with a good response to IVF-ET inhibited significantly P production, and blocking IgG were detected in 85% women with a good response to IVF-ET. Conclusion(s): This study identified FSH-blocking IgG in a high proportion of women with a good response to IVF-ET. The significance of this remains questionnable. (Fertil [email protected] 1998;69:46-52. 01998 by American Society for Reproductive Medicine.) Key Words: Premature ovarian failure, in vitro fertilization, granulosa cell culture, ovarian antibodies, FSH-blocking IgG Premature ovarian failure (POF) is characterized by the cessation of ovarian function after puberty and before age 40 years. This condition is easily demonstrated by secondary amenorrhea, estrogen deficiency, and elevated circulating levels of both FSH and LH. Its prevalence is < 1% among women of reproductive age, Infertile women with regular menses, normal or moderately elevated FSH levels,

and impaired superovulatory response during IVF-ET programs might represent an early stage of POF referred as “occult ovarian failure” in previous studies (1,2). Both conditions may represent a continuum of the same disease leading to premature menopause (3). The involvement of autoimmunity in POF and failure to IVF-ET is suggested by several lines of evidence. POF is widely observed con-

currently with other autoimmune disorders such as Hashimoto’s thyroiditis, Grave’s disease (4), Addison’s disease (4), or myasthenia gravis (5). Lymphocytic infiltration of the ovaries is described in POF (6), as well as alterations of Tand B-cell functions (7). Antibodies to ovarian tissue in POF are raised against granulosa cells, theta cells, luteal cells, and oocytes (8). Antibodies to human ovary and gonadotropins are also detected in women with failure to IVF-ET (1, 9, 10). Such antibodies could be induced by repeated hormonal stimulation and follicular microtrauma during IVF-ET cycles, as suggested by previous studies (11, 12). In PGF and women with a poor response to IVF-ET, ovarian failure to exogenous gonadotropin stimulation could result from antibodies against gonadotropins or their receptors. This hypothesis was tested in POF with conflicting results (13-15). In the present work, a study was conducted on 14 POF and 43 infertile women submitted to IVF-ET to identify serum antibodies directed against FSH or its receptor and able to block the steroidogenic action of recombinant human FSH on human granulosa cells in culture.

MATERIALS AND METHODS Clinical Characteristics of Patients and Controls POF Patients

with a good response to IVF-ET included 14 women whose E, concentration reached >2,000 pg/mL in response to ovarian stimulation. Twenty-three of the 29 women with a poor response to IVF-ET and 8 of 14 women with a good response to IVF-ET had previously undergone IVF-ET, and all patients from both groups had been submitted to repeated ovarian stimulation by exogenous gonadotropins before IVF-ET stimulation The cause of infertility was tubal blockade in one-half of the patients from both groups, and chronic dysovulation, male factor, or unexplained in the remaining, with no difference in the distribution between women with a poor response to IVF-ET and women with a good response to IVF-ET. Frequency of pelvic-abdominal surgery, endometriosis, and number of previous IVF was the same in both groups (Table 1). Controls

The control group consisted of 22 healthy women matched for age (range, 30-40 years) and recruited within the Endocrinology unit staff. All were fertile women, 8 were taking oral contraceptive agents, 9 had an intrauterine device, and 5 used external contraceptive devices. None had received hormonal treatment for infertility. Informed consent was obtained from all controls, and Institutional Review Board approval was obtained.

Fourteen women with POF who had bad amenorrhea before the age of 40 years and an FSH concentration of >40 mIU/mL were selected retrospectively. One patient had secondary amenorrhea for 3 years with moderate elevation of FSH level (23 mIU/mL), but very low E, levels at several determinations (12-20 pg/mL). All patients had secondary amenorrhea at a median age of 3 1 years (range, 17-40 years), and their median age at study was 34 years (range, 22-42 years). All patients had a normal chromosomal karyotype. An associated autoimmune disease was observed in 6 of 14 women with POF, and an ovarian biopsy was performed in 5 of 14 women with POF, 2 of 5 being afollicular and 3 of 5 showing numerous immature follicles suggestive of the resistance ovary syndrome.

In all patients, samples were drawn in the fasting state in basal condition, during the early or late follicular phase, before ovarian stimulation in infertile women submitted to IVF-ET. In controls, samples were collected during the follicular phase in 15 of 22 women, and during the luteal phase in the remaining. Sera were stored at -20°C. LH and FSH were measured by immunoradiometric assay (bioMCrieux, Marcy l’Etoile, France). Assay sensitivity for LH and FSH was 0.3 mIU/mL and 0,15 mIU/mL, respectively. Plasma E, was measured by RIA (bioMerieux), and assay sensitivity was 10 pg/mL.

Infertile Women Undergoing IVF-ET

Purification of IgG

Forty-three infertile women who were candidates for an IVF-ET program were selected retrospectively for a 2-year period. Most patients (33 of 43) were enrolled in a GnRH agonist (GnRH-a)-hMG-hCG trial for ovulation induction, and 10 patients were enrolled in an hMG/hCG trial without prior pituitary desensitization by GnRH-a. Infertile patients undergoing IVF-ET were thus separated into two groups based on their maximum E, concentration in response to ovarian stimulation: women with a poor response to IVF-ET included 29 women who failed to reach an E, concentration of 500 pg/mL in response to ovarian stimulation, and women

One-step purification of IgG was performed using protein A-Sepharose CL4B chromatography. One milliliter of serum was passed on a protein A column (0.75 g dry beads per column), preequilibrated with 0.02 mol/L PBS buffer pH 7.2. Absorbed immunoglobulins were eluted with 0.05 mol/L glycin buffer pH 3. The eluates were dialyzed against phosphate-buffered saline (PBS), and further against minimum essential medium (MEM) before storage at -80°C. IgG recovery ranged from 4.4 to 9.6 mg/mL of serum, whereas IgA and IgM recovery were CO.1 mg/mL and 0.01 mg/mL, respectively.

FERTILITY

& [email protected]

Determination of Immunoreactive FSH, LH, and E, in Plasma Samples

47

Clinical characteristics

of infertile patients undergoing

Variable

IVF-ET.

Women with a good response to IVF-ET (n = 14)

Women with a poor response to IVF-ET (n = 29)

P value

33 2 4 25 ? 5

35 _’4 27 ? 5

0.15 NS* 0.20 NS*

13129 7129 1129 8129 20129 7129 1.4 2 1.5 312%

0.75 NSt 0.69 NSt 0.24 NSt 1.00 NSt 0.23 NS$ 0.69 NSt 0.13 NS$

Mean age (? SD) at study (yr) Mean age (? SD) at onset of infertility (yr) Cause of infertility (no. with condition/total No. of women) Tubal blockade Chronic dysovulation Male factor Unexplained Pelvic-abdominal surgery Endometriosis Mean no. (-+ SD) of previous IVF attempts No. with endocrine disease/total no. of women

7114 2114 2114 4114 7114 2114 2.2 2 1.8 2/141/

Note. NS = not significant. * Determined by Student’s r-test. t Determined by Yates’ 2 test. $ Determined by the 2 test. 8 Determined by Spearman’s test. ((Two patients had hyperprolactinemia. ‘pTwo patients had hyperprolactinemia, and one had idiopathic hirsutism.

Bioassay of Antibodies Antagonizing FSH Activity To investigate a blocking effect of IgG on FSH action, a human granulosa cell assay was used. This assay was based on the ability of FSH to stimulate P production in human granulosa cells human granulosa cells were obtained from periovulatory follicles aspirated in ovaries of patients with tubal factor infertility or male factor infertility and undergoing IVF-ET treatment. Human granulosa cells were separated from follicular fluid (FF) by centrifugation. Human granulosa cells were separated from red blood cells on Ficoll plaque gradient. Cells were plated (lo5 cells/200 pL) in MEM + 10% fetal calf serum (FCS) under humidified 5% CO, atmosphere at 37°C. At day 2, culture mediums were removed and cells were cultured in MEM +5% FCS for another 48 hours. At day 4, culture mediums were removed and cells were cultured for 24 hours in 200 ~_LL of MEM 0.5% bovine serum albumin, containing recombinant human FSH (Serono Laboratories, Boulongne, France) with or without addition of IgG (5, 50, and 500 pg/mL) from patients and controls. Human granulosa cells from follicles obtained from several women were pooled for each experiment, and in each experiment IgG from POF, infertile women undergoing IVF, and controls were attempted. In all experiments, each incubation was performed in triplicate samples, and P measurement was performed by RIA in duplicate samples. In the culture conditions described above, preliminary experiments were performed to determine the range of the P response to human FSH (Fig. 1). We observed that human 48

Reznik et al.

Ovarian failure and FSH-blocking

IgG

granulosa cells exhibited a dose-dependent increase in P production in response to human FSH. At the submaximal dose (0.1 mIU/mL) of human FSH, a twofold increase of P production above control was observed and this concentration was chosen to detect an IgG blocking effect on human FSH-induced steroidogenesis. Each experiment was thus validated when a stimulatory FSH response was actually obtained from pooled human granulosa cells cultured in the absence of IgG. Results were first normalized in nanograms of P released by 100,000 granulosa cells. To discard the between-patient variation, values for each patient were expressed as a ratio of the mean response to IgG plus recombinant human FSH versus the mean response to sole recombinant human FSH, which account the within-patient variation (sum of the between IgG +/- variation and of the residual variation). A value >2 SD below the mean from control women was considered a significant inhibitory response.

Statistical Analysis The analysis was performed with a statistical software Statview (Abacus Concepts, Inc., Berkeley, CA, 1994), on a power Macintosh 7200/75. All results were expressed as the means -+ SD. Between-group comparison was performed with the appropriate test: ,$ and Yate’s test for categorical variables, and Student’s t-test, Spearman test, and one-way analysis of variance (ANOVA) for continuous variables. The validity of each ANOVA was assessed by Levene test for homogeneity of variances. In the absence of homogeneity, a Kruskal-Wallis test was performed. P CO.05 was defined as statistically significant. Vol. 69, No. 1, January 1998

Standard curve of P response to recombinant human FSH in human granulosa cells. Human granulosa cells (1O5cells/200 pL) were incubated with increasing doses of recombinant human FSH, and extracellular P was measured 24 hours later in culture medium. Each point represents the mean +- SEM of three experiments performed in triplicate.

Recombinant hFSH (W/L)

RESULTS Clinical and Hormonal Characteristics of POF, Infertile Women, and Controls When women with a poor response to IVF-ET and women with a good response to IVF-ET were compared, women did not differ significantly according to the age at onset of infertility, the cause of infertility, the frequency of pelvic surgery or endometriosis, and the number of previous ovarian stimulation for IVF-ET (Table 1). Mean baseline FSH (mIU/mL) and LH (mIU/mL) concentrations in women with POF (mean 2 SD, 79 2 59 and 41 + 24) were significantly higher (P < 0.0001) than in women with a poor response to IVF-ET (9 + 5 and 7.4 -+ 4.8), than in women with a good response to IVF-ET (5.5 -C 1.9 and 7.5 k 3.9), and than in control women (6.4 + 3.4 and 6.6 + 5). Mean E, concentration (pg/mL) was significantly lower (P < 0.05) in women with POF (23 2 22, mean ? SD) than in women with a poor response to IVF-ET (100 t 48), women with a good response to IVF-ET (115 5 SS), and control women (106 + 116). Women with a poor response to IVF-ET and women with a good response to IVF-ET did not have significantly different basal FSH, LH, and E, concentrations. In contrast, the E, peak and the number of oocytes obtained by ovulation induction (mean +- SD) were significantly higher (P < 0.0001) FERTILITY & [email protected]

in women with a good response to IVF-ET (3470 2 550 and 18 z!z6, respectively) than in women with a poor response to IVF-ET (210 + 135 and 6 k 2).

Effect of IgG From POF on Recombinant Human FSH-Stimulated P Production Immunoglobulin G from the 22 controls had no effect on recombinant human FSH-stimulated P production, whatever the IgG concentration in the culture medium (Fig. 2). The mean P response from human granulosa cells exposed to IgG from POF was not significantly different from the mean P response to IgG from controls, at IgG concentrations of 5 pg/mL (99 t 18% versus 107 + 16%), 50 pg/mL (102 k 24% versus 104 + 12%), and 500 I.Lg/mL (93 + 24% versus 94 + 14%). Nevertheless, at a concentration of 50 pg/mL, IgG from 3 of 14 POF induced an inhibition of P response to recombinant human FSH >2 SD below the mean response obtained with IgG from controls (Fig. 2, middle panel), reflecting an IgG inhibitory effect on FSH action in these three POF patients. An inhibitory effect was also observed from one patient at an IgG concentration of 5 pg/mL (Fig 2, upper panel) and from two patients at an IgG concentration of 500 pg/mL (Fig. 2, lower panel). It is of interest that two of these POF with blocking IgG exhibited at ovarian biopsy an histologic aspect of abundant immature follicles arrested in the primordial stage of development, which suggests an ovarian resistance to the action of gonadotropins.

Effect of IgG From Infertile Women Undergoing IVF-ET on Recombinant Human FSH-Stimulated P Production The mean P response of human granulosa cells hGE exposed to IgG from women with a poor response was not significantly different from the mean P response to IgG from controls at IgG concentrations of 5 pg/rnL (106 ? 24% versus 107 + 16%), 50 pg/mL (102 + 20% versus 104 I+_ 12%), and 500 pg/mL (91 +- 15% versus 94 + 14%). At the IgG concentrations of 5 pg/mL and 50 Fg/mL, only 2 of 29 women with a poor response to IVF-ET had FSH-blocking IgG as defined above (Fig. 2, upper and middle panels). In contrast, the mean P response of human granulosa cells exposed to IgG from women with a good response to IVF-ET was significantly lower than the response to IgG from women with a poor response to IVF-ET and from controls at IgG concentrations of 50 pg/mL (86 + 21% versus 102 + 20% and 104 2 12%; P < 0.05) and 500 pg/mL (60 2 24% versus 91 k 15% and 94 2 14%; P < 0.0001). An ordinal interaction between IgG concentration and percent inhibition of P response was observed in women with a good response to IVF-ET versus controls, which was statistically significant by ANOVA for two factor experiments with repeated measures on one factor (P -=c 0.01). These results demonstrate a dose-response effect of IgG on 49

Individual P responses of human granulosa cells incubated for 24 hours with recombinant human FSH and IgG from patients and controls. Progesterone secretion represents the ratio of FSH-stimulated P production in the presence of IgG to FSH-stimulated P production without the presence of IgG. Horizontal lines represent ? 2 SD of the mean response of controls. POF, premature ovarian failure.

IgG were identified in 6 of 14 women with a good response to IVF-ET (43%) at IgG concentration of 50 pg/mL and 12 of 14 women with a good response to IVF-ET (85%) at IgG concentration of 500 pg/mL (Fig. 2, middle and lower panels). No correlation was observed between the magnitude of the FSH-blocking activity of IgG or the number of IVF-ET cycles, the cause of infertility, or a history of pelvic surgery or endometriosis.

DISCUSSION In the present study, we tested for the presence of antibodies directed against FSH or its receptor in infertile women who underwent IVF-ET cycles and in women with POF. We assessed the hypothesis that such antibodies might be present in these various clinical conditions as pathogenic antibodies possibly involved in the mechanism of ovarian deficiency, or at least as a marker of an autoimmune process. To detect the presence of FSH action-blocking IgG, we used human granulosa cells cultured in the presence of submaximal doses of human FSH. Previously, it was described that FSH could stimulate P as well as E, production in human granulosa cells from periovulatory follicles (16); therefore, we used the measurement of P production as a criteria for FSH action on human granulosa cells. Our key finding was that women with a poor response to IVF-ET exhibited a low prevalence of FSH activity-blocking IgG (7%) as did patients with POF (20%), suggesting that such antibodies may not explain the ovarian deficiency observed in both groups. In contrast, women with a good response to ovarian stimulation exhibited a prevalence as high as 85% of FSH activity-blocking IgG. Few studies were designed to detect in the sera of infertile women undergoing IVF-ET cycles antibodies against gonadotropins or their receptors. Meyer et al. (9) found a high prevalence of antibodies to LH or FSH in women with a poor response to IVF-ET as determined by a homologous ELISA. Moncayo et al. (10) demonstrated the presence of autoantibodies binding in vitro the LH/hCG receptor from corpora lutea of bovine ovaries in some infertile women undergoing IVF-ET. These autoantibodies were also detected in patients with endometriosis and polyendocrinopathy.

COMdS

POF

Poor

Iwpoase

Good response

FSH-stimulated P response of human granulosa cells in women with a good response to IVF-ET. The FSH-blocking 50

Reznik et al.

Ovarian

failure and FSH-blocking IgG

In POF, the ovary is thought to be the target of an autoimmune process and ovarian autoantibodies were detected in the sera of patients with POF (8, 17). The pathogenicity of such ovarian antibodies is suggested [l] by the increased risk of POF in women with autoimmune polyglandular disease and preserved ovarian function when steroidal cell antibodies are detected in their sera (18); [2] by the demonstration of a cytotoxic effect of sera from patients with Addison’s disease and POF on human granulosa cells (19); and [3] by case reports of normalized menses or pregnancy outcome after immunosuppressive therapy in infertile paVol. 69, No. 1, January 1998

tients or POF patients with ovarian antibodies in their sera (6,20,21).

which islet cell antibodies disappear within 1 year after clinical onset of the disease (25).

In approximately 15% of women with POF, normal follicles may be present at ovarian biopsy, with a characteristic aspect of primordial follicles arrested at an early stage of development, suggesting a failure of high gonadotropin concentrations to stimulate follicular growth (22). This state was referred to as the resistant ovary syndrome, and in 1982 Chiauzzi et al. (5) first demonstrated an FSH-binding inhibitory activity in the IgG fraction of two women with concurrent POF and myasthenia gravis.

Finally, our results indicate the unexpected presence of FSH-blocking IgG in those patients with the best response to gonadotropin stimulation. The mechanism of this IgG production and its significance remain unclear. At least, we can conclude that these data do not support a pathogenic role of FSH-blocking IgG in the lack of response to ovulation induction observed in some infertile women submitted to IVF-ET cycles. Nevertheless, such antibodies could represent a marker of the different stages of an autoimmune process implicated in infertility and POF, leading to progressive ovarian failure and autoantibodies disappearance. Such a hypothesis needs to be evaluated in further studies.

The search for antibodies to gonadotropin receptors gave conflicting results. Austin et al. (13) failed to identify antibodies with LH receptor-binding inhibitory activity, whereas Van Weissenbruch et al. (14) found a prevalence of 81% FSH blocking IgG in 26 POF. However, the interpretation of experiments using FSH receptors from rat testis (5) or rat ovaries (14) to identify antibodies to human FSH receptors is questionnable, as species specificity of human FSH binding to its receptor was clearly demonstrated in recent studies using recombinant human FSH receptors (23). A recent study by Anasti et al. (15) using human recombinant FSH and LH receptors failed to detect inhibitory antibodies in 38 POF. Our results using an homologous system with human granulosa cells and recombinant human FSH showed FSH activity-blocking IgG in only 3 of 14 POF, associated in two patients with the histologic changes suggestive of the resistant ovary syndrome. In the present study, our results question the clinical significance of FSH-blocking IgG. Such FSH-blocking IgG identified in women with a good response to IVF-ET could possibly be directed to FSH receptor or alternatively to FSH itself. Such antibodies could result from repeated hormonal stimulation (1 l), microtrauma of the follicles (12), or pelvic surgery (17). The possibility that FSHblocking IgG, as identified in our study, could increase FSH, nevertheless, would not explain the different prevalence of such FSH-blocking IgG in women with a good response to IVF-ET versus women with a poor response to IVF-ET. Moreover, antibodies to FSH should cross-react with FSH measurement and likely induce false low FSH concentrations by IRMA determination, which were not observed in any serum. Alternatively, the increase in FSH-blocking IgG in women with a good response to IVF-ET could be the consequence of a marked antigenic stimulation of the immune system at an early stage of an autoimmune process as seen in prediabetes, which corresponds to the period preceding the development of type 1 diabetes mellitus (24). The absence of such FSH-blocking IgG observed in women with a poor response to IVF-ET and POF could be attributable to poor folliculogenesis and therefore, low antigenic triggering. This latter condition could represent a later stage of the autoimmune disease, as observed in clinical diabetes mellitus in FERTILITY & [email protected]

The authors are indebted to Francette Pottier for her excellent technical assistance, to Pierre Leymarie for his useful criticism in reviewing the manuscript, and to Serono Laboratories (Boulogne, France) for providing human recombinant FSH.

Acknowledgment:

References 1. Cameron IT, O’Shea FC, Rolland JM, Hughes EG, de Kretser DM, Healy DL. Occult ovarian failure: A syndrome of infertility, regular menses, and elevated follicle-stimulating hormone conc&trat~ons. J Clin Endocrinol Metab 1988;67:1190-4. 2. Ahmed Ebbiary NA, Lenton EA, Salt C, Ward AM, Cooke ID. The significance of elevated basal follicle stimulating hormone in regularly menstruating infertile women. Hum Reprod 1994;9:245-52. 3. Rebar RW, Silva de Sa MF. The reproductive age: Premature ovarian failure. In: Serra G, editor. Comprehensive endocrinology, the ovary. New York Raven Press, 1982:399-446. 4. Labarbera A, Miller M, Ober C, Rebar R. Autoimmune etiology in premature ovarian failure. Am J Reprod Immunol 1988;16:115-21. 5. Chiauzzi V, Cigorraga S, Escobar ME, Rivarola MA, Charreau EH. Inhibition of follicle-stimulating hormone receptor binding by circulating immunoglobulins. J Clin Endocrinol Metab 1982;54: 1221-8. 6. Coulam CB, Kempers RD, Randall RV. Premature ovarian failure: Evidence for an autoimmune mechanism. Fertil Steril 1981:36:23840. 7. Pekonen F, Siegberg R, Makinen T, Miettinen A, Yli-Korkala 0. Immunological disturbances in patients with premature ovarian failure. Clin Endocrinol 1986:25:1-6. 8. Damewood MD, Zacur HA, Hoffman GJ, Rock JA. Circulating antiovarian antibodies in premature ovarian failure. Obstet Gynecol 1986; 68:850-4. 9. Meyer WR, Lavy G, de Chemey AH, Visintin I, Economy K, Luborsky JL. Evidence of gonadal and gonadotropin antibodies in women with a suboptimal ovarian response to exogenous gonadotropin. Obstet Gynecol 1990;75:795-9. 10. Moncavo H, Moncavo R, Benz R. Wolf A. Lauritzen CH. Ovarian failure
14. Van Weissenbruch MM, Hoek A, Van Vliet-bleeker I, Schoemaker J, Drexhage H. Evidence for existence of immunoglobulins that block ovarian granulosa cell growth in vitro. A putative role in resistant ovary syndrome? J Clin Endocrinol Metab 1991;73:360-7. 15. Anasti JN, Flack MR, Froehlich J, Nelson LM. The use of human recombinant gonadotropin receptors to search for immunoglobulin G-

51

16. 17.

mediated premature ovarian failure. J Clin Endocrinol Metab 1995;SO: 824-8. Bar-Ami S. Gitav-Goren H. Altered steroidoaenic activitv of human granulosa-lutein cells at different cell densit& culture. Mel Cell Endocrinol 1993;90:157-64. Luborsky JL, Visintin I, Boyers S, Asari T, Caldwell B, de Cherney A. Ovarian antibodies detected by immobilized antigen immunoassay in patients with premature ovarian failure. J Clin Endocrinol Metab 1990; lo~6o-75 ._.__ Ahonen P. Miettinen A. Perheentuua J. Adrenal and steroidal cell antibodies ‘in patients with autoimmune polyglandular disease type 1 and risk of adrenocortical and ovarian failure. J Clin Endocrinol Metab 1987;64:494-500. McNattv, KP. Short RV. Barnes EW. Irvine WJ. The cvtotoxic effect of serum from patients with Addison’s disease and autoimmune ovarian failure on human granulosa cells in culture. Clin Exp Immunol 1975; 22:378-84. Bateman BG, Nunley WC, Kitchin JD III. Reversal of apparent pre-

21. 22.

II.

18.

19.

20.

52

Reznik et al.

Ovarian failure and FSH-blocking

IgG

23.

24. 25.

mature ovarian failure in a patient with myasthenia gravis. Fertil Steril 1983;39:108-10. Taylor R, Smith NM, Angus B, Home CHW, Dunlop W. Return of fertility after twelve years of autoimmune ovarian failure. Clin Endocrinol 1989;31:305-8. Jones GS, de Moraes-Ruehsen M. A new syndrome of amenorrhea in association with hypergonadotropism and apparently normal ovarian follicular apparatus. Am J Obstet Gynecol 1969;104:597600. Tillv JL. Aihara T. Nishimori K. Jia XC, Billie H. Kowalski KI. et al. Expression of recombinant human follicle-stir&lating hormone receptor: species-specific ligand binding, signal transduction and identification of multiple ovarian messenger ribonucleic acid transcripts. Endocrinoloev 1992:131:799-806. Eisenb& GS.‘Type 1 diabetes mellitus: A chronic autoimmune disease. N Engl J Med 1986;314:1360-8. Lendrum R, Walker G, Gamble DR. Islet cell antibodies in iuvenile diabetes mellitus of recent onset. Lancet 1975;1:880-2.

Vol. 69, No. 1, January 1998