Faculty Bibliography

We optimized storage conditions and validated a sensitive immunofluorometric assay (IFMA) for urinary gonadotropins. Assay linearity and parallelism for luteinizing hormone (LH) and follicle-stimulating hormone (FSH) was observed to 0.04 IU/L. Urinary LH and FSH were unaffected by changes of osmolarity from 0.5 to 3.0 mOsm/kg, and from pH 4.5 to 10.5. Serum and urine measurements of the hormones correlated well over a wide range of values: for LH, R2 = 0.94, P < 0.01; for FSH, R2 = 0.71, P < 0.01 (n = 304). Preservation of urine with glycerol (70 mL/L) and storage at -20 degrees C yielded > 80% recovery of LH and FSH after 51 weeks; this was comparable with recovery for acetone extracts of urine. Untreated urine showed loss of activity by 4 weeks of storage. Preserving urine specimens with glycerol is a simple method of storage for longitudinal study and compares favorably with acetone extraction. IFMAs can measure urinary gonadotropins reproducibly over a wide range of pH and osmotic conditions.

To determine whether cortisol has an effect on hypothalamic-pituitary-gonadal function, we studied 11 eumenorrheic women in the early follicular phase of consecutive menstrual cycles by performing daytime 10-min blood sampling, one before and one during hydrocortisone administration. Daily blood sampling for gonadotropins and sex steroids was also performed. LH pulsations were determined by modification of a widely used threshold method and compared by paired t-testing. The LH interpulse interval was significantly prolonged (95 +/- 5 to 119 +/- 14 min; p = 0.001), and the mean LH pulse amplitude remained unchanged (1.3 +/- 0.1 and 1.5 +/- 0.2 mIU/ml) with glucocorticoid exposure. Mean estradiol was not altered (46 +/- 5 and 43 +/- 3 pg/ml), but mean LH and FSH from pooled serum aliquots were slightly but significantly reduced (2.6 +/- 0.2 to 2.2 +/- 0.2, 5.5 +/- 0.4 to 4.5 +/- 0.3 mIU/ml; p = 0.004, 0.012, respectively). Mean progesterone levels were also decreased (0.8 +/- 0.3 to 0.5 +/- 0.2 ng/ml; p = 0.011) in the presence of exogenous glucocorticoid. Twenty-four-hour urinary free cortisol levels confirmed a substantial increase in free cortisol excretion (74 +/- 10 to 305 +/- 50 micrograms/day; p = 0.001). These results demonstrate that cortisol can slow LH pulse frequency and, by inference, hypothalamic GnRH secretion, in a manner that appears independent of corticotropin releasing factor. Excess cortisol alone may therefore play a role in the development of stress-associated menstrual disturbances.

OBJECTIVE: To determine the effect of human chorionic gonadotropin (hCG) on relaxin secretion by long-term cultures of luteinized human granulosa cells (GC). DESIGN: Luteinized human GC were collected from 10 women undergoing in vitro fertilization (IVF) cycles. Luteinized human GC from each woman were plated in replicate wells at 1 x 10(5) cells/well and exposed to medium 199 (GIBCO, Grand Island, NY), medium 199 with 1 IU/mL hCG, and/or medium 199 with 100 IU hCG/mL. Luteinized human GC were maintained for up to 40 days in culture. Spent media were changed every 2 days and assayed for relaxin and progesterone (P) at the conclusion of each experiment. SETTING: Tertiary care center. PATIENTS, PARTICIPANTS: Luteinized human GC were obtained from women undergoing controlled ovarian hyperstimulation for IVF with one of the following regimens: (1) clomiphene citrate with human menopausal gonadotropins (hMG); (2) hMG alone; or (3) hMG with leuprolide acetate. All women were less than 40 years of age, in good health, and were not taking medications other than those used in the ovulation-induction regimen. MAIN OUTCOME MEASURES: Levels of P and relaxin in spent media. RESULTS: Relaxin secretion by luteinized human GC was dependent on hCG stimulation and was detected only after a time lag in culture. After relaxin secretion was detected, it was maintained throughout the culture period (10 to 22 days). Luteinized human GC produced P immediately under both basal and stimulated conditions. Progesterone production continued throughout the culture period with hCG-stimulated cells producing significantly greater P after 4 to 8 days in culture. CONCLUSIONS: Luteinized human GC obtained at the time of oocyte retrieval secrete relaxin in response to hCG stimulation and secrete P under both basal and hCG-stimulated conditions, thereby serving as a model to explore luteal function and control.

Insulin participates in regulating ovarian function in normal and in pathological states. This effect of insulin may be mediated by ovarian insulin receptors. We have previously characterized human ovarian insulin receptors and began to examine their regulation in vitro. The present study examines regulation of human ovarian insulin receptors in vivo. Stromal ovarian tissue was obtained from 21 women during an indicated surgical procedure. Ten women were premenopausal and 11 were postmenopausal. Specific 125I-insulin binding to stromal ovarian fragments ranged from 2.5% to 7.3%/mg protein. 125I-insulin binding to stromal fragments correlated positively with 125I-insulin binding to circulating leucocytes (r = .57; P less than .01). When postmenopausal and premenopausal women were analyzed separately, this relationship persisted in postmenopausal women (r = .70; P less than .05), but not premenopausal women. 125I-insulin binding to stromal ovarian fragments correlated negatively with age (r = -.63; P = .005). 125I-insulin binding to stromal ovarian fragments tended to correlate negatively with plasma insulin levels in postmenopausal women (r = -.67; P = .06), but not in premenopausal women. Plasma insulin levels correlated negatively with serum SHBG (r = -.62; P = .003). The percent free testosterone levels correlated positively with plasma insulin levels in premenopausal women (r = .95; P = .0001), but not in postmenopausal women.(ABSTRACT TRUNCATED AT 250 WORDS)

Methionine-auxotrophic L1210 cells were used to study the effect of methotrexate (MTX) on methionine uptake and metabolism. MTX was shown to inhibit amino acid transport systems and cause a decrease of methionine uptake into L1210 cells. Conversely, a nonmetabolizable amino acid analogue reduced MTX uptake into L1210 cells. MTX also blocked the transfer of the beta carbon from serine into methionine. Therefore, methionine deprivation may be an additional mechanism of action for MTX in methionine-auxotrophic tumor cells.