Increased peripheral T lymphocyte activation in patients with karyotypically normal spontaneous premature ovarian failure

Transcription

1 FERTILITY AND STERILITY Copyright Cl 1991 The American Fertility Society Printed on acid-free paper in U.S.A. Increased peripheral T lymphocyte activation in patients ith karyotypically normal spontaneous premature ovarian failure Larence M. Nelson, M.D.*t Lorene M. Kimzey, R.N.,B.S.:!: George R. Merriam, M.D.* Thomas A. Fleisher, M.D. Developmental Endocrinology Branch, National Institute of Child Health and Human Development; Nursing Department, Clinical Center, National Institutes of Health; and Immunology Service, Clinical Center, National Institutes of Health, Bethesda, Maryland Objective: To determine if soluble interleukin 2 (IL-2) receptor measured in serum by an enzymelinked immunosorbent assay (ELISA) might be useful in managing patients ith karyotypically normal spontaneous premature ovarian failure. Design: Prospective, controlled observation. Setting: Tertiary care research institution. Interventions: None. Patients, Participants: Tenty-four patients ith karyotypically normal spontaneous premature ovarian failure comprised the study group. Forty-to healthy men and omen comprised the normal reference group. Main Outcome Measures: We measured peripheral T lymphocyte human leukocyte antigen locus DR (HLA-DR) expression and IL-2 receptor expression using monoclonal antibodies and flo cytometry. We measured soluble IL-2 receptor levels in serum using an ELISA. Results: Consistent ith previous findings, our patients had significantly higher HLA-DR expression on peripheral T lymphocytes (5.3 ±.46) as compared ith controls (3.5 ±.34) (mean ± SEM, P <.1). Seven patients also had elevated IL-2 receptor expression on peripheral T lymphocytes (P <.5). Hoever, soluble IL-2 receptor levels in the serum did not differ significantly from normals. Conclusions: Patients ith karyotypically normal spontaneous premature ovarian failure have a modest increase in peripheral T lymphocyte activation measured by flo cytometry. This degree of activation does not result in increased soluble IL-2 receptor release measured by ELISA. Fertil Steril 55:182, 1991 Autoimmune ovarian failure may occur alone, or as part of a constellation of autoimmune endocrine Received October 5, 199; revised and accepted February 1, * Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health. t Reprint requests: Larence M. Nelson, M.D., Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Building 1, Room 1N262, Bethesda, Maryland :j: Nursing Department, Clinical Center, National Institutes of Health. Immunology Service, Clinical Center, National Institutes of Health. failure syndromes. 1 Ovarian histopathology in some patients ith ovarian failure has shon developing follicles infiltrated ith lymphocytes and plasma cells. 2-4 In the past, interest in autoimmune premature ovarian failure has focused primarily on ovarian autoantibodies, but groing interest has developed regarding the role that cellular immunity plays in this. 5-9 Rabinoe et al. lo have shon that as a group, patients ith premature ovarian failure have increased T lymphocyte activation evidenced by increased human leukocyte antigen locus DR (HLA-DR) expression on peripheral T lymphocytes. Human leukocyte antigen locus DR is a class II major histocompatibility complex antigen and is a marker of T lymphocyte activation. 182 Nelson et al. Lymphocyte activation in ovarian failure Fertility and Sterility

2 Activated T lymphocytes, in addition to expressing HLA-DR on their cell surface, also generate the 55-kd alpha chain of the interleukin-2 (ll-2) receptor. Not only do activated T lymphocytes display this IL-2 receptor on their cell surface, but they also release into the blood a soluble form of this IL-2 receptor that can be measured by enzyme-linked immunosorbent assay (ELISA)Y A rapid and reproducible method to detect increased immune activation, such as an ELISA, might prove useful to clinicians caring for patients ith ovarian failure. Therefore, e measured serum soluble IL-2 receptor levels by ELISA in patients ith karyotypic ally normal spontaneous premature ovarian failure to determine if this test can detect immune activation in these patients. To confirm that indeed our patient group had increased T lymphocyte activation as predicted by a previous report,lo e also measured T lymphocyte HLA-DR and IL-2 receptor (alpha chain) expression by flo cytometry, a more complex method. MATERIALS AND METHODS Subjects and Clinical Parameters By letter or notice in medical journals, e recruited patients ith spontaneous premature ovarian failure for ongoing clinical trials evaluating treatments for this condition. To qualify for referral, patients ere to be <4 years old ith at least 4 months of amenorrhea and to have had serum follicle-stimulating hormone values > 4 IU /L on at least to occasions, at least 1 month apart. Karyotypes ere obtained on all patients. Patients ith abnormal karyotypes or evidence to suggest iatrogenic ovarian failure ere not included. Tenty-four omen ith premature ovarian failure and a median age of 32 years (range 15 to 39) comprised the study group. The patients had the ovarian failure diagnosed at a median age of 3 years (range 14 to 36), and the median time since diagnosis as 2 years (range.25 to 12). To omen (8%) had primary amenorrhea. Telve omen (5%) had previously been pregnant, and 8 omen (33%) had previously given birth. Six patients (25%) had hypothyroidism, 1 patient (4%) had Addison's disease, 1 patient (4%) had Raynaud's syndrome. Six patients ith undiagnosed prior illnesses had the folloing problems: 1 patient had a severe normochromic normocytic anemia during pregnancy and required several transfusions; 2 patients had prior transient pain, selling, and erythema of finger joints that had not been evaluated; and 3 patients had prior prolonged febrile illnesses that required hospitalization and led to no definitive diagnosis. Control subjects ere healthy men and omen studied to establish normal adult reference ranges. They ere therefore not matched for age or gender (n = 16 for HLA-DR and IL-2 receptor by flo cytometry, median age 37, range 24 to 59 years) (n = 26 for soluble IL-2 receptor assay, median age 36, range 2 to 66 years). T Lymphocyte HLA-DR and IL-2 Receptor Expression We measured T lymphocyte HLA-DR and IL-2 receptor (alpha chain) expression as part of a comprehensive immunophenotyping panel. We separated mononuclear blood cells by density gradient and stained them ith directly conjugated (fluorescein isothiocyanate or phycoerythrin) monoclonal antibodies (Coulter Corporation, Hialeah, FL; Becton-Dickinson Immunocytometry Systems, San Jose, CA). We measured dual color immunofluorescence using a FACScan (Becton-Dickinson Immunocytometry Systems) and gated lymphocytes based on forard versus sidelight scatter. We confirmed the validity of the lymphocyte gate using CD45 and CD14 (Leucogate; Becton-Dickinson Immunocytometry Systems). The percent of total lymphocytes in each subpopulation identified as generated for patients and controls, and the absolute number of each cell subpopulation as then calculated. Soluble IL-2 Receptor by ELISA We measured soluble IL-2 receptor levels by methods previously described. 12 Briefly, e used a sandich ELISA ith to different monoclonal antibodies to the alpha chain of the IL-2 receptor. We used alkaline phosphatase-conjugated antibody and para-nitrophenyl phosphate (Sigma Chemical Company, St. Louis, MO) as the enzyme-substrate reaction yielding measurable color. The degree of substrate conversion as determined at 45 nm ith a Titertek ELISA reader (Flo Laboratories, McLean, VA). To convert absorbance values to a corresponding concentration of soluble IL-2 receptor expressed as U/mL, e used a standard curve generated from a knon standard source of soluble IL-2 receptor. 12 Autoantibodies We measured antinuclear antibodies (ANA) by indirect immunofluorescence using Hep-2 substrate, rheumatoid factor by latex agglutination, and rapid plasma reagin (RPR) using standard methods. An- Nelson et al. Lymphocyte activation in ovarian failure 183

3 Table 1 Summary Statistics for Peripheral T Lymphocyte HLA-DR and IL-2 Receptor Expression and for Serum Soluble IL-2 Receptor Levels in Patients and Controls Statistics Probability HLA-DR Controls 3.5 ±.34 (.9 to 6.1)b Patients 5.3 ±.46 <.1 IL-2 receptor Controls 3.5 ±.25 (1.5 to 5.5) Patients 4.6 ±.62 >.5 Soluble IL-2 receptor C (U/ml) Controls 35 ± 1.7 (16 to 58) Patients 284 ± 1.1 >.5 Expressed as percent oflymphocytes positive for this marker. b Values are means ± SEM ith normal ranges in parentheses. C Log-normal distribution; see text for statistical methods. tithyroglobulin antibodies and antithyroid microsomal antibodies ere performed by indirect hemagglutination using standard methods. Antiparietal cell antibodies ere measured by indirect immunofluorescence using rat stomach (SmithKline Beecham Clinical Laboratories, Van Nuys, CA). Antiovarian and antisteroid cell immunohistochemical studies used rabbit ovary and human testis, respectively, and ere performed by the Department of Pathology, University of Florida, College of Medicine, Gainesville, Florida. 13 Statistical Analysis We analyzed data using the Bright Stat-Pack 14 on the DEC-lO mainframe computer at the National Institutes of Health, Bethesda, Maryland. Summary statistics for normally distributed data ere calculated using standard statistical methods. For lognormally distributed data, e calculated summary statistics by using the natural logarithm of values to estimate the mean, SD, and confidence interval (CI) and then reported findings using the antilogarithm of the calculated values. 15 We used Wilcoxon's to-sample rank sum statistic and Fisher's exact test to evaluate differences beteen patients and controls. We set P <.5 for significance. RESULTS Tests for autoimmunity in the 24 patients ith karyotypically normal spontaneous premature ovarian failure shoed the folloing: 9 patients (38%) had an ANA 1:4, 5 patients (21%) had antithyroid microsomal antibodies 1:1, 2 patients (8%) had antithyroglobulin antibodies 1: 1, 4 patients (17%) had antiparietal cell antibodies 1:1, and 1 patient (4%) had rheumatoid factor positive at 1:32. No patient had a positive RPR. The 1 patient ith knon Addison's disease demonstrated antisteroid cell antibodies, but no other patient demonstrated antisteroid cell antibodies, and no patient demonstrated antiovarian antibodies. Sixteen patients (67%) demonstrated a positive result on at least one of these tests for autoantibodies. In the control group, results for HLA-DR and IL-2 receptor expression on peripheral T lymphocytes ere normally distributed. The results for soluble IL-2 receptor levels ere log-normally distributed. Summary statistics are shon in Table 1. Findings are represented graphically in Figures 1 and 2. Patients ith karyotypically normal spontaneous premature ovarian failure had significantly higher HLA-DR expression on peripheral T lymphocytes (CD3+) as compared ith controls (P <.1), and seven patients (29%) had HLA-DR expression above the normal range. Although the mean IL-2 receptor expression in patients ith ovarian failure as not significantly different from controls, values in seven patients (29%) ere clearly above the normal range for IL-2 receptor expression on peripheral T lymphocytes (P <.5). The data are presented as percentage of total lymphocytes; the findings are comparable hen expressed as absolute number of T lymphocytes positive for these activation markers. The geometric mean serum soluble IL-2 receptor levels ere not significantly different from control values; to patients and to controls ere above the normal range. Overall, in 12 patients (5%) e found evidence for increased T lymphocyte activation based on either elevated T lymphocyte HLA-DR expression, T lymphocyte IL-2 receptor expression, or soluble IL-2 receptor levels in serum. Nine of these 12 patients ere positive for ANA, antithyroid antibodies, or antiparietal cell antibodies (Table 2). When findings ere segregated on the basis of evidence for T lymphocyte activation and on the basis of clinical or laboratory evidence for another autoimmune, patients ith past history or laboratory findings to suggest the presence of another autoimmune did not have a significantly increased likelihood of peripheral T lymphocyte activation hen compared ith patients ithout such findings. Hoever, of 12 patients ith no clinical evidence for another autoimmune, 8 (67%) had evidence for increased T lymphocyte activation by at least one of the three measures. 184 Nelson et ai. Lymphocyte activation in ovarian failure Fertility and Sterility

4 14 a:: 12 U 1 a::-...!:s.!.a:: -e 8 s (/)::1:!Be 8 a:: Z ILe( 4 U 2 o IL-2 RECEPTOR HLA-DR I::) I::) - I ---- fl I I CONTROLS PATIENTS CONTROLS PATIENTS IL-2 RECEPTOR AND HLA-DR EXPRESSION Figure 1 Percent of peripheral T lymphocytes expressing IL-2 receptor and HLA-DR in patients and controls. The horizontallines represent the respective means. The rectangles represent the 95% CI for controls. DISCUSSION We found significantly increased HLA-DR expression on peripheral T lymphocytes in 29% of patients ith karyotypic ally normal spontaneous premature ovarian failure as compared ith normal reference range subjects, hich is in agreement ith the findings of Rabinoe et al. 1 Using monoclonal antibody-defined T lymphocyte phenotyping, they found 35% of patients ith premature ovarian failure had elevated HLA-DR expression on peripheral T lymphocytes. In our study, e extend the evidence for increased immune activation by using another measure of T lymphocyte activation-expression of the alpha chain of the IL-2 receptor on the surface of activated T lymphocytes. Of our patients ith karyotypically normal spontaneous premature ovarian failure, 29% had increased IL-2 receptor expression on peripheral T lymphocytes. Serum soluble IL-2 receptor levels are elevated in patients ith autoimmune s such as rheumatoid arthritis and systemic lupus erythematosus. 19,2 Soluble IL-2 receptor levels reflect increased T lymphocyte activation 12 and correlate ith disease activity in these s. 16-2o Patients ith type I diabetes mellitus, an autoimmune destruction of pancreatic islets, also have elevated soluble IL-2 receptor levels 21 in the serum and increased T lymphocyte activation evidenced by T lymphocyte HLA-DR expression. 22 By analogy, patients ith premature ovarian failure and increased T lymphocyte activation (evidenced by elevated HLA-DR expression) might also be expected to demonstrate elevated soluble IL-2 receptor serum levels. - Hoever, measurement of soluble IL-2 receptor levels in our patients did not demonstrate significantly increased T lymphocyte activation. There are several possible explanations for this. It may reflect the more localized nature of the autoimmune activity in premature ovarian failure, hich involves less tissue mass than other, more generalized autoimmune s. Nevertheless, in type I diabetes mellitus, hich is also a rather localized autoimmune process, one report has demonstrated elevated soluble IL-2 receptors. 21 This difference might be explained by histologic evidence that suggests that the autoimmune process in ovarian failure may be even more limited than in diabetes mellitus. In some cases autoimmune premature ovarian failure spares primordial follicles, and the process attacks only developing graafian follicles. 2-4 When used to detect increased immune activation in our patients, serum soluble IL-2 receptor level measured by ELISA appears to be less sensitive than monoclonal antibodydefined T lymphocyte phenotyping. Also, ithout direct biopsy confirmation of an autoimmune process, the precise etiology of ovarian failure in our patients remains unproven, although both the evidence for increased T lymphocyte activation and the presence of at least one other autoantibody in 67% of our patients suggest an alteration in immune activation. None of our patients had antiovarian antibodies detectable by immunofluorescence. These findings suggest that the antiovarian antibody test used here has lo sensitivity. This is in general agreement ith Rabinoe et al. 1 ho, using indirect immunofluorescence on guinea pig ovary, found antiovarian antibodies in only 2 of a group of 23 premature ovarian failure patients ith 1 8..J 8... E '""CJ ::J 8 a:: 4 + l- ll W U a:: J III 4- ::J..J (/) 1 CONTROLS PATIENTS Figure 2 Serum soluble IL-2 receptor levels in patients and controls. The horizontal lines represent the respective geometric means. The rectangle represents the 95% CI for controls. Nelson et al. Lymphocyte activation in ovarian failure 185

5 Table 2 Clinical Features of Patients With Increased T Lymphocyte Activation Evidence for T Antithyroid antibody titer lymphocyte ANA Antiparietai Patient activation titer Microsomal Thyroglobulin cell titer Clinical findings 1 HLA-DR (8.3%)a 2 IL-2 receptor 1:32 (7.1%)b 3 HLA-DR (6.2%) IL-2 receptor (9.7%) 4 HLA-DR 1:32 (7.%) 5 IL-2 receptor 1:4 (6.2%) Soluble IL-2 receptor (64 U/mL)' 6 HLA-DR (11.9%) IL-2 receptor (9.8%) 7 HLA-DR 1:16 (1.2%) IL-2 receptor (1.9%) 8 HLA-DR (6.5%) 9 HLA-DR 1:4 (6.3%) 1 IL-2 receptor (8.6%) 11 HLA-DR 1:1 (6.7%) 12 Soluble IL-2 receptor 1:4 1:1 (868 U/mL) a Normal range.9% to 6.1 % of T lymphocytes. b Normal range 1.5% to 5.5% of T lymphocytes. 1:4 Thyroid function normal 1:1 Hypothyroidism, psoriasis Hypothyroidism Unexplained severe anemia during pregnancy Psoriasis; prior prolonged enigmatic febrile illness requiring steroid therapy 1:2 1:16 Hypothyroidism, psoriasis 'Normal range 16 to 58 U/mL. evidence for increased T lymphocyte activation. These to patients also had antiadrenal antibodies, suggesting the possibility that even in these patients they may have measured cross-reacting antisteroid cell antibodies rather than ovarian specific antibodies.23 Thus, to independent reports no demonstrate a lo prevalence of antiovarian antibodies in premature ovarian failure patients ho as a group have increased peripheral T lymphocyte activation. These findings do not support the usefulness of antiovarian antibody tests that employ subprimate animal tissue. Patients ith premature ovarian failure may experience spontaneous or therapy-associated pregnancies,24.25 but no proven therapy exists that has been subjected to controlled prospective trial. To develop a rational therapy for ovarian failure, the mechanisms causing follicular dysfunction must be clearly defined. This ill require sensitive and specific tests that are able to identify those patients experiencing ovarian failure because of a selective process impairing normal graafian follicle development and function but sparing primordial follicles. We have shon that measuring serum levels of soluble IL-2 receptor fails to detect increased peripheral T lymphocyte activation in patients ith karyotypically normal spontaneous premature ovarian failure, and thus this test is unlikely to be useful in managing these patients. Patients ith premature ovarian failure as a group have increased T lymphocyte activation as determined by HLA DR expression. lo Hoever, this T lymphocyte activation has not been demonstrated to be specific to antiovarian autoimmune activity. Consequently, it ould be premature to assign a role to lymphocyte phenotyping in the clinical evaluation of premature 186 Nelson et al. Lymphocyte activation in ovarian failure Fertility and Sterility

6 ovarian failure, especially considering its expense. Furthermore, tests hich use subprimate tissue to search for antiovarian antibodies appear to have lo sensitivity. A sensitive and specific test that reflects autoimmune activity against graafian follicles is needed. Ackrledgments. Weare indebted to Ms. Barbara Filmore of the Developmental Endocrinology Branch for technical expertise; to Marcella Proch, B.S.N., and the 1 West and 9th floor Ambulatory Care Research Facility nursing staff of the Clinical Center, National Institutes of Health for invaluable help; to Carole C. Kurman, B.S., and David L. Nelson, M.D., of the Metabolism Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, for assistance in determining soluble IL-2 receptor levels; and to Noel K. Maclaren, M.D., Department of Pathology and Laboratory Medicine, University of Florida College of Medicine, Gainesville, Florida, for immunohistochemical studies. REFERENCES 1. Neufeld M, MacLaren NK, Blizzard RM: To types of autoimmune Addison's disease associated ith different polyglandular autoimmune (PGA) syndromes. Medicine (Baltimore) 6:355, Wolfe CDA, Stirling RW: Premature menopause associated ith autoimmune oophoritis. Case report. Br J Obstet Gynaecol 95:63, Case Records of the Massachusetts General Hospital (Case ). N Engl J Med 317:127, Gloor E, Hurlimann J: Autoimmune oophoritis. Am J Clin Pathol 81:15, Friedman CI, Neff J, Kim MH: Immunologic parameters in premature follicular depletion: T and B lymphocytes, T-cell sub-populations, cutaneous reactivity, and serum immunoglobulin concentrations. Diagn Immunol 2:48, Pekonen F, Siegberg R, Makinen T, Miettinen A, Yli-Korkala : Immunological disturbances in patients ith premature ovarian failure. Clin Endocrinol (Oxf) 25:1, Ho PC, Tang GW, Fu KH, Fan MC, Laton JW: Immunologic studies in patients ith premature ovarian failure. Obstet Gynecol 71:622, Mignot MH, Drexhage HA, Kleingeld M, Van de Plassche Boers, EM, Rao BR, Schoemaker J: Premature ovarian failure. II: considerations of cellular immunity defects. Eur J Obstet Gynecol Reprod Bioi 3:67, Hill JA, Welch WR, Faris HMP, Anderson DJ: Induction of class II major histocompatibility complex antigen expression in human granulosa cells by interferon gamma: a potential mechanism contributing to autoimmune ovarian failure. Am J Obstet Gynecol 162:534, Rabinoe SL, Ravnikar VA, Dib SA, George KL, Dluhy RG: Premature menopause: monoclonal antibody defined T lymphocyte abnormalities and antiovarian antibodies. Fertil Steril 51:45, Nelson DL, Rubin LA, Kurman CC, Fritz ME, Boutin B: An analysis of the cellular requirements for the production of soluble interleukin-2 receptors in vitro. J Clin Immunol 6: 114, Rubin LA, Kurman CC, Fritz ME, Biddison WE, Boytin B, Yarchoan R, Nelson DL: Soluble interleukin-2 receptors are released by activated lymphocytes in vitro. J Immunol 135: 3172, Elder M, Maclaren N, Riley W: Gonadal autoantibodies in patients ith hypogonadism and/or Addison's disease. J Clin Endocrinol Metab 52:1137, Cole BR, Rodbard D, Munson PJ: Bright Stat-Pack. Bethesda, Maryland, National Institutes of Health, Hines WGS: Geometric mean. In Encyclopedia of Statistical Sciences, Vol. 3, Edited by S Kotz, NL Johnson, CB Read. Ne York, John Wiley and Sons, Inc., 1983, p Symons JA, Wood NC, Di Giovine FS, Duff GW: Soluble IL-2 receptor in rheumatoid arthritis. J Immunol 141:2612, Wood NC, Symons JA, Duff GW: Serum interleukin-2-receptor in rheumatoid arthritis: a prognostic indicator of disease activity? J Autoimmun 1:353, Campen DH, Horitz DA, Quismorio FP, Ehresmann GR, Martin WJ: Serum levels of interleukin-2 receptor and activity of rheumatic diseases characterized by immune system activation. Arthritis Rheum 31:1358, Sememzato G, Bambara LM, Biasi D, Frigo A, Vinante F, Zuppini B, Trentin L, Feruglio C, Chilosi M, Pizzolo G: Increased serum levels of soluble interleukin-2 receptor in patients ith systemic lupus erythematosus and rheumatoid arthritis. J Clin ImmunoI8:447, Manoussakis MN, Papadopoulos GK, Drosos AA, Moutsopoulos HM: Soluble interleukin 2 receptor molecules in the serum of patients ith autoimmune diseases. Clin Immunol Immunopathol 5:321, Giordano C, Galluzzo A, Marco A, Panto F, Amato MP, Caruso C, Bompiani GD: Increased soluble interleukin-2 receptor levels in the sera oftype 1 diabetic patients. Diabetes Res 8:135, Jackson RA, Morris MA, Haynes BF, Eisenbarth GS: Increased circulating la-bearing T cells in type I diabetes mellitus. N Engl J Med 36:785, Sotsiou F, Bottazzo GF, Doniach D: Immunofluorescence studies on autoantibodies to steroid-producing cells, and to germline cells in endocrine disease and infertility. Clin Exp ImmunoI39:97, Szlachter BN, Nachtigall LE, Epstein J, Young BK, Weiss G: Premature menopause: a reversible entity? Obstet Gynecol 54:396, Shapiro AG, Rubin A: Spontaneous pregnancy in association ith hypergonadotropic ovarian failure. Fertil Steril 28:5, 1977 Nelson et al. Lymphocyte activation in ovarian failure 187