Replacement of dehydroepiandrosterone enhances T-lymphocyte insulin binding in postmenopausal women*†

M.nopause FERTILITY AND STERILITY Copyright

©

Vol. 63, No, 5, May 1995

1995 American Society for Reproductive Medicine

Printed on acid-free paper in U. S. A

Replacement of dehydroepiandrosterone enhances T-lymphocyte insulin binding in postmenopausal women*t

Peter R. Casson, M.D.t§ Lisa C. Faquin, Ph.D.11 Frankie B. Stentz, Ph.D.** Arthur B. Straughn, Pharm.D. tt

Richard N. Andersen, Ph.D.t Guy E. Abraham, M.D.tt John E. Buster, M.D.t§§

University of Tennessee, Memphis, Memphis, Tennessee, and Optimox Corporation, Torrance, California

Objective: To demonstrate bioavailability of 3 weeks of oral micronized DHEA and to delineate changes induced on insulin sensitivity, morphometric indexes, and lipoprotein profiles. Design: Oral micronized DHEA (50 mg/d) was administered in 3-week treatments to 11 postmenopausal women in a prospective, placebo-controlled, randomized, blinded, crossover trial with an interarm washout. After dose (23 hour) serum DHEA, DHEAS, T, and cortisol levels were measured, as were fasting lipoproteins, oral glucose tolerance tests (OGTT), Tlymphocyte insulin binding and degradation, and urine collagen cross-links. Morphometric changes were determined by hydrostatic weighing. Results: Dehydroepiandrosterone sulfate, DHEA, T, and free T increased up to two times premenopausal levels with treatment. Fasting triglycerides declined; no change in collagen cross-links or morphometric indexes was noted. Oral glucose tolerance test parameters did not change, but both T-Iymphocyte insulin binding and degradation increased with DHEA. Conclusion: Fifty milligrams per day of oral DHEA gives supraphysiologic androgen levels; 25 mg/d may be more appropriate. Dehydroepiandrosterone enhanced tissue insulin sensitivity and lowered serum triglycerides. Rationale is provided for postmenopausal replacement therapy with this androgen. Fertil Steril 1995;63:1027-31 Key Words: DHEA replacement, insulin sensitivity, menopause

The adrenal androgens DHEA and DHEAS decline with age, independent of cortisol secretion (1), Received February 17, 1994; revised and accepted November 25,1994. * Presented in part at the conjoint meeting of The American Fertility Society and the Canadian Fertility and Andrology Society, October 11 to 14, 1993, Montreal, Quebec, Canada. t Supported in part by The American College of Obstetrics and Gynecology, Ortho Academic Training Fellowship, of which Dr. Casson is the 1990-1991 recipient, and by clinical research center grant, United States Public Health Services RR00211-27, at the University of Tennessee, Memphis, Tennessee. t Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, University of Tennessee, Memphis. § Reprint requests and present address: Peter R. Casson, M.D., Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030 (FAX: 713-798-8431). Vol. 63, No.5, May 1995

in a pattern analogous to the estrogen deficiency of menopause. This decline is particularly relevant to human aging because of mounting evidence linking these steroids to potentiation of antioncogenic (2, 3) and immune processes (4, 5), control of obesity (6, 7), and bone homeostasis (8). Lipoprotein profiles (7), the incidence of cardiovascular disease (9), and insulin action (10, 11) also may be affected beneficially by these hormones. These benefits, and the

II Present address: College of Human Sciences, Florida State University, Gainesville, Florida. ** Clinical Research Center, Department of Medicine, University of Tennessee, Memphis. tt School of Pharmacy, University of Tennessee, Memphis. H Optimox Corporation. §§ Present address: Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas.

Casson et al. DHEA enhances lymphocyte insulin binding

1027

observation that the age-related deficiency of adrenal androgens is accelerated by ovarian failure (12), make the concept of adrenal androgen replacement in older women appealing. Attempts to replace DHEA in humans, using high doses and oral administration, have been complicated by supraphysiologic serum levels of the hormone and its metabolites and subsequent deleterious androgenic bioconversion (13). We previously demonstrated that administration of 50 mg/d micronized oral DHEA more closely approximates the adrenal androgen milieu of a premenopausal woman, while minimizing androgenic bioconversion (14). In another part of the present study, we demonstrated that 3 weeks of this dose of oral micronized DHEA resulted in enhancement of natural killer cell cytotoxicity and number (5). In that protocol, we also examined the bioavailability and physiologic impact ofthis low dose of DHEA on body composition, bone homeostasis, lipoproteins, and insulin sensitivity in postmenopausal women, all of which may be targets for beneficial DHEA effects. We hypothesized that DHEA replacement will result in restitution of the adrenal androgen milieu in these women, with enhanced insulin sensitivity, beneficial changes in lipoprotein profiles, decreased bone turnover, and increased lean body mass. We now report the results of this study. MATERIALS AND METHODS Human Subjects

Mter institutional review board approval, 11 postmenopausal women were recruited. Menopause (> 1 year without spontaneous menses) was confirmed by a single serum FSH > 40 mIU/mL (conversion factor to SI unit, 1.00). Previous estrogen replacement was discontinued for 2:3 weeks before the protocol. Inclusion criteria included normal or greater than normal weight for height (defined by a body mass index> 20 kg/m2), serum DHEAS < 125 JIg/dL (conversion factor to SI unit, 0.02714), and no current medications. The subjects' screening history and physical had to be normal as well as their complete blood count, electrolytes, thyroid, liver, renal function tests, and fasting serum insulin levels. Baseline screening T and E2 levels were within the normal ranges. Subject characteristics have been published previously (5) and have been reproduced in Table 1. Study Design

In a prospective, randomized, double-blind, crossover trial, subjects were given either a tablet containing 50 mg of micronized DHEA (in a wax vegeta1028

Table 1 Subject Screening Characteristics (n = 11) Mean Age (y) Weight (kg) Body mass index (kg/m2) FSH (mIU/mLlt DHEAS (J.lgldl)t E2 (pglmL)§ T (ngldL)11

56.1 71.1 26.1 84 75 41 54

Range 45 53.6 21.2 44 34 22 6

to to to to to to to

66 103.8 37.5 100 121 59 98

* From Casson PR, Andersen RN, Herrod HG, Stentz FB, Straughn AB, Abraham GE, et al. (5). Reprinted by permission of the publisher. t Conversion factor to SI units, 1.00. t Conversion factor to SI units, 0.02714. § Conversion factor to SI units, 3.671. II Conversion factor to SI units, 3.467.

ble oil matrix with a silica-based excipient) or an identical placebo, each morning at 8:00 A.M. for 3 weeks. A 2-week wash out was followed by a crossover to the other arm. Blood for hormone measurements and insulin binding studies was drawn at baseline and 23 hours after the last dose. Serum was stored at - 20°C until assayed. Hormone Measurements

Serum DHEA, DHEAS, T, and cortisol (F) were measured at the beginning and end of each treatment arm. Dehydroepiandrosterone was assayed with a tritiated extraction RIA (Wein Scientific, Succasanna, NY). Dehydroepiandrosterone sulfate, F, E2 (Pantex Ltd., Santa Monica, CA) and T (lCN Ltd., Costa Mesa, CA) were determined with direct assays using 1251. All samples from each subject were run in the same assay for each steroid. The intra-assay coefficients of variation were :::;9% for all steroids measured. The blind was not broken until all assays were performed and the results were available to the investigators. At the beginning and end of each treatment period, albumin was measured by a commercial spectrophotometric method (Kodak Ektachem, Rochester, NY). Sex hormone-binding globulin (SHBG) was determined by a saturation assay (15). The levels of albumin, SHBG, and total T were then used to calculate a value offree T by the method of Wild et al. (16). A standard 75 mg oral glucose tolerance test (OGTT; measuring serum insulin and glucose at 0, 30, 60, 90, 120, and 180 minutes after administration) was performed, and a 24-hour urine sample was collected for pyridinoline and deoxypyridinoline measurements, by a previously validated procedure using high-performance liquid chromatography (17) (Nichols Institute, San Juan Capistrano, CA). Standard fasting lipid profiles, were done

Casson et al. DHEA enhances lymphocyte insulin binding

Fertility and Sterility

using an enzymatic rate procedure (Beckman Astra system, Brea, CA). Lean body mass and percentage fat were determined by hydrostatic weighing using a submersion tank after correction for lung residual volume, vital capacity, body temperature, and ambient pressure (18). At the beginning and end of each treatment period, blood samples were drawn for in vitro studies of T -lymphocyte insulin binding and degradation as described previously (19). Monocytes were separated on a Ficoll-Hypaque gradient and were cultured with 1 p,g/mL phytohemaglutin in 25-mL flasks at 10 X 106 cells/mL for 3 days in 5% CO 2, at 37°C, to give an activated T-Iymphocyte cell suspension. These cells were then centrifuged, resuspended, and cell viability was confirmed to be >95% by Trypan blue exclusion. Specific insulin binding was determined by incubating the cells with 187,000 cpm 125I-insulin in binding buffer for 2 hours at 15°C. Nonspecific binding was determined by a similar incubation with the addition of 105 ng/mL unlabeled insulin. Specific binding was calculated as total 125I-insulin binding minus nonspecific binding and was expressed as a percentage of total count. The cells then were washed to remove non-ceIl-associated radioactivity for the insulin degradation studies, resuspended in buffer, and incubated at 37°C for 60 minutes. Aliquots were removed, added to trichloroacetic acid, centrifuged, and the supernatant was separated from the precipitate. Insulin degradation was determined by counting the radioactivity in the supernatant and the precipitate (19). Cell-specific insulin degradation was calculated by the ratio of radioactivity in the supernatant and precipitated fractions and was expressed as a percentage of total count. Data Analysis

A split plot design was used. Significant differences in the dependent variables during treatment and placebo periods were compared by analysis of variance with repeated measures, after ensuring no time effect existed. RESULTS Hormone Measurements

Significant increases over placebo were seen for the day 21 DHEAS, DHEA, T, and free T levels. These levels were equivalent to or somewhat above established (mean ± SD) premenopausal values (530 ± 248 ng/dL for DHEA [20], 206 ± 86 p,g/dL for DHEA-S [21], and 20 to 50 ng/mL for T [22], depending on menstrual phase). Sex hormone-binding globulin declined significantly (Table 2). There was Vol. 63, No.5, May 1995

a significant elevation from baseline in serum F levels in the placebo group (4.18 ± 1.17 ng/dL) compared with the treatment group (0.65 ± 1.84 ng/ dL). However, 24-hour free urinary F levels did not change with treatment or placebo. Lipoproteins, Body Mass Indexes, and Bone Turnover

Serum triglycerides declined significantly (P 0.02) compared with placebo (Table 3), but fasting serum high-density lipoprotein, low-density lipoprotein, and total cholesterol levels did not change. There was no change in 24-hour urinary pyridinoline or deoxypyridinoline excretion over the treatment or placebo periods (Table 3). =

Oral Glucose Tolerance Tests and T-Lymphocyte Insulin Binding

T-Iymphocyte insulin binding and degradation increased significantly after DHEA treatment as compared with placebo (Fig. 1A and B). Moreover, these changes in binding and degradation were seen in 11 of 11 subjects. No significant change in lean body mass, percentage body fat, or mean insulin or glucose area under the curve after OGTT were seen with DHEA administration (Table 3). DISCUSSION

In this preliminary study 50 mg/d oral DHEA was given to postmenopausal women, elevating their day 21 DHEA, DHEAS, and T levels significantly over placebo. The serum levels of these three steroids increased to or somewhat above established premenopausal norms. It appears, therefore, as these are nadir values, that a dose of 50 mg/d of this preparation of micronized DHEA may be excessive for replacement, and some of the beneficial effects of normalized DHEA and DHEAS may have been masked by excessive biochemical androgenization. Thus, although no differences in clinical insulin resistance or morphometric indexes was seen in this study, a lower dose (perhaps 25 mg/d) or a longer duration of administration may be needed for discernment of significant changes. Surprisingly, F levels were elevated during placebo treatment, whereas no change was observed with DHEA administration. However, free urinary F levels remained unchanged. Similarly, urinary excretion of collagen cross-links (pyridinoline and deoxypyridinoline), measured as a parameter of bone resorption (17), did not change with 3 weeks of DHEA treatment. Given the limited time frame of drug administration, this finding was not

Casson et al. DHEA enhances lymphocyte insulin binding

1029

Table 2

Hormone Measurements Before (Day 0) and After (Day 21) Treatment* DHEASt

Drug

Placebo

Drug

81.2 :+: 9.6 69.9:+: 8.6

166.7:+: 23.9 418.5:+: 29.811

74.2:+: 8.5 456.3:+: 71.511

Placebo

Drug

160.9 :+: 27.8 186.6:+: 22.7

25.0:+: 3.4 74.2:+: 7.611

ng/dL

l'l!/dL

Day 0 Day 21

T§

DHEAt

Placebo

Drug

23.6 :+: 3.0 26.5:+: 2.5

60.6:+: 9.8 46.5:+: 7.3~

Drug

55.6 :+: 8.7 60.9:+: 8.6

0.25:+: 0.03 0.93:+: 0.1311

Placebo ng/dL

0.25 :+: 0.03 0.26:+: 0.03

§ Conversion factor to SI units, 3.467.

t Conversion factor to SI units, 0.0274.

liP < 0.0001.

:j: Conversion factor to SI units, 0.03467.

~

unexpected. However, adrenal androgens appear to playa role in postmenopausal osteoporosis (8) and evaluation of parameters of bone turnover in further longer-term studies of DHEA administration still may be worthwhile. Enhancement in T-Iymphocyte insulin binding and degradation, a previously defined marker of insulin sensitivity (23), is a finding in accord with other less rigorous studies of DHEA effect. Intravenous 17-hour infusions of DHEA enhance postreceptor insulin sensitivity in women (11), and 1 month of DHEA treatment in a noninsulin-dependent diabetic woman resulted in a 30% improvement in multiple parameters of insulin sensitivity (24). With longer administration of a lower dose, this enhanced sensitivity to insulin on the cellular level may have translated into clinically detectable enhancement ofinsulin sensitivity. Although an OGTT represents a reasonable initial screen for clinical insulin resistance (25), it lacks sensitivity. More rigorous in vivo evaluation, such as an intravenous tolbutamide augmented glucose tolerance test or hyperinsulinemic

Placebo nM

ng/dL

* Values are means:+: SEM.

Free T§

SHBGII

P

=

0.001.

euglycemic clamp (25) may have detected changes that OGTTs may overlook. This study provides evidence that oral replacement of DHEA in adrenal androgen deficiency states, including advanced age, is a valid concept. Postmenopausal women are particularly good candidates for adrenal androgen replacement therapy as menopause appears to accelerate the rate of diminution of adrenal androgen secretion (12), and many ofthe putative effects of these compounds may be complementary to estrogen replacement therapy. This may allow for reduction of the dose of estrogen used with consequent reduction of side effects (such as vaginal bleeding) and, as a result, increased compliance. In summary, the results from this study, presented here and elsewhere (5), demonstrate [1] the optimal dose of this preparation of micronized oral DHEA for postmenopausal replacement may be <50 mg/d; [2] DHEA enhances tissue insulin sensitivity and, perhaps with longer-term administration oflower doses, may enhance insulin effect in vivo; and [3] DHEA administration decreases serum triglycerides.

Table 3 Changes From Baseline Seen in Postmenopausal Women Over a 3-Week Course of Treatment With DHEA or Placebo for Various End Point Parameters DHEA* (n Fasting lipoproteins (mg/dL):j: High-density lipoprotein Low-density lipoprotein Total cholesterol Triglycerides Bone turnover 24-Hour urinary pyridinoline (pmollJ.lmol creatinine) 24-Hour urinary deoxypyridinoline (pmollJ.lmol creatinine) Morphometric indexes Lean body weight (kg) Body fat (%) Weight (kg) Insulin sensitivity Post-OGTT insulin area under the curve (J.lU . mLlh) Post-OGTT glucose area under the curve (mg· dLlh)

~4.9

Placebo* (n

~3.0

~38.12 ~

2.8 3.9 2.6 14.6§

4.6 1.4

~

4.4 1.8

~2.6

~

~0.5

~

0.18

~

0.17 0.24 0.51

~0.20 ~

~

~ ~9.00 ~

3.5

~

~0.04 ~

0.02 30.7 2.2

* Values are means ~ SEM. t Values are ranges. 1030

= 11)

Casson et al. DHEA enhances lymphocyte insulin binding

~

~ ~

28.7 5.2

2.7 2.5 13.73

0.33 0.35

~ ~ ~

~

~ ~

~18.8

~

17.0

~

Baselinet (n

= 11)

2.4 7.1 7.1 8.5

33 61 154 41

3.8 2.1

26.4 to 59.4 7.0 to 31.6

0.21 0.28 0.53

33.7 to 55.7 29.1 to 45.9 53.5 to 101.0

27.2 14.9

to to to to

= 11)

76 168 270 485

76.0 to 290.8 225.8 to 534.09

:j: Conversion factor to SI units, 0.02586. § P = 0.02 compared with placebo change.

Fertility and Sterility

A

p < O.OO()1

8. •

IlHEA

!TIl

PlACEBO

INSULIN

BINDING (%total count)

9.

10. DAV21

DAY 1

B

DAY 1

DAY 21

p
11. •

DHEA

ll:il

PLACEBO

12. INSUUN DEGRADAnON (%totalcount)

13.

DAY 1

DAY 21

DAY 1

14.

DAY 21

Figure 1 (A), T-Iymphocyte insulin binding in postmenopausal women after 3 weeks of DHEA or placebo treatment (mean ± SEM, n = 11). (B), T-Iymphocyte insulin degradation in postmenopausal women in response to 3 weeks of DHEA or placebo treatment (mean ± SEM, n = 11).

15.

16. Acknowledgments. The authors acknowledge Mr. Charles Hakala of Belmar Pharmaceuticals for supplying the DHEA preparation and Elaine Bryant, Ph.D., for her invaluable statistical help. We also acknowledge the technical assistance of Ms. Alice Milem, Mr. Ed Umstot, Ms. Karen Rufus, and Mr. Jeff Hollis. Finally, the assistance of Ms. Brooke Tate and Ms. Peggy Tate in preparing the manuscript was appreciated greatly.

17.

18.

REFERENCES 1. Parker LN, Odell WD. Decline of adrenal androgen production measured by radioimmunoassay of urinary unconjugated dehydroepiandrosterone. J Clin Endocrinol Metab 1978;47: 600-2. 2. Schwartz AG, Pashko L, Whitcomb JM. Inhibition of tumor development by dehydroepiandrosterone and related steroids. Toxicol Pathol 1986; 14:357 -62. 3. Gordon GB, Shantz LM, Talalay P. Modulation of growth differentiation and carcinogenesis by dehydroepiandrosterone. Adv Enzyme ReguI1987;26:355-82 4. Suzuki T, Suzuki N, Daynes RA, Engleman EG. Dehydroepiandrosterone enhances IL-2 production and cytotoxic effector function of human T cells. Clin Immunol Immunopathol 1991;61:202-11. 5. Casson PR, Andersen RN, Herrod HG, Stentz FB, Straughn AB, Abraham GE, et al. Oral dehydroepiandrosterone in physiologic doses modulates immune function in postmenopausal women. Am J Obstet Gynecol 1993; 169:1536-9. 6. MacEwen EG, Kurzman ID. Obesity in the dog: role of the adrenal steroid dehydroepiandrosterone (DHEA). J Nutr 1991; 121:S51-5. 7. Nestler JE, Barlascini CO, Clore IN, Blackard WG. Dehydroepiandrosterone reduces serum low density lipoprotein Vol. 63, No.5, May 1995

19.

20.

21. 22.

23.

24.

25.

levels and body fat but does not alter insulin sensitivity in normal men. J Clin Endocrinol Metab 1988;66:57-61. Taelman P, Kaufman JM, Janssens X, Vermeulen A. Persistence of increased bone resorption and possible role of dehydroepiandrosterone as a bone metabolism determinant in osteoporotic women in late postmenopause. Maturitas 1989; 11:65-73. Barrett-Connor E, Khaw KT, Yen SSC. A prospective study of dehydroepiandrosterone sulfate, mortality, and cardiovascular disease. N Engl J Med 1986;315:1519-24. Schriock ED, Buffington CK, Hubert GD, Kurtz BR, Kitabchi AE, Buster JE, et al. Divergent correlations of circulating dehydroepiandrosterone sulfate and testosterone with insulin levels and insulin receptor binding. J Clin Endocrinol Metab 1988;66:1329-31. Schriock ED, Buffington CK, Givens JR, Buster JE. Enhanced post-receptor insulin effects in women following DHEA infusion. J Soc Gynecol Invest 1994;1:74-8. Cumming DC, Rebar RW, Hopper BR, Yen SSC. Evidence for an influence of the ovary on circulating DHEA-S levels. J Clin Endocrinol Metab 1982;54:1069-71. Mortola JF, Yenn SSC. The effects of oral DHEA on endocrine-metabolic parameters in post menopausal women. J Clin Endocrinol Metab 1990; 71:696-704. Buster JE, Casson PR, Straughn AB, Dale D, Umstot ES, Chiamori N, et al. Postmenopausal steroid replacement with micronized dehydroepiandrosterone (DHEA): preliminary oral bioavailability and dose proportionality studies. Am J Obstet Gynecol 1992; 166:1163-70. Rosner W. A simplified method for the quantitative determination of testosterone-estradiol-binding globulin activity in human plasma. J Clin Endocrinol Metab 1972;34:983-8. Wild RA, Umstot ES, Ranney GB, Andersen RN, Givens JR. Androgen parameters and their correlation with body weight in one hundred thirty-eight women thought to have hyperandrogenism. Am J Obstet Gynecol 1983; 146:602-6. Uebelhart D, Schlemmer A, Johansen JS, Gineyts E, Christiansen C, Delmas PD. Effect of menopause and hormone replacement therapy on the urinary excretion of pyridinium cross-links. J Clin Endocrinol Metab 1991; 72:367 -73. Wilmore JH. The use of predicted and consistent residual volume in the assessment of body composition by underwater weighing. Med Sci Sports Exerc 1968; 1:87 -90. Buffington CK, EI-Shiekh T, Kitabchi AE, Matteri R. Phytohemagglutinin (PHA) activated human T-Iymphocytes: concomitant appearance of insulin binding, degradation and insulin-mediated activation of pyruvate dehydrogenase (PDH). Biochem Biophys Res Commun 1986; 134:412-9. Abraham GE, Buster JE, Kyle FW, Corrals PC, Teller RC. Radioimmunoassay of plasma pregnenolone, 17-hydroxypregnenolone and dehydroepiandosterone under various physiological conditions. J Clin Endocrinol 1973; 37:140-4. Simon J, Buster J. Plasma DHEA-S concentration of normal ovulatory women. Clin Chern 1983;29:1319-20. Abraham GE. Ovarian and adrenal contributions to peripheral androgens during the menstrual cycle. J Clin Endocrinol Metab 1974;39:340-6. Buffington CK, Givens JR, Kitabchi AE. Opposing actions of dehydropiandrosterone and testosterone on insulin sensitivity; in vivo and in vitro studies of hyperandrogenic females. Diabetes 1991;40:693-700. Buffington CK, Pourmotabbed G, Kitabchi AE. Amelioration of insulin resistance in diabetes with dehydroepiandrosterone. Am J Med Sci 1993;306:320-4. Bergman RN, Finegood DT, Ader M. Assessment of insulin sensitivity in vivo. Endocr Rev 1985;6:45-86.

Casson et aI. DHEA enhances lymphocyte insulin binding

1031