Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
0021-972X/98/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1998 by The Endocrine Society Vol. 83, No. 2 Printed in U.S.A. Identification of Progesterone Receptor in Human Subcutaneous Adipose Tissue SUSAN N. O’BRIEN, BRENDA H. WELTER, KIMBERLY A. MANTZKE, THOMAS M. PRICE AND Greenville Hospital System/Clemson University Biomedical Cooperative Research Program, Clemson University, Clemson, South Carolina 29634 tissue of premenopausal women. These preliminary studies revealed that PR messenger RNA levels are higher in the stromal-vascular fraction as opposed to the adipocyte fraction. Western blot analysis demonstrates both PR protein isoforms (human PR-A and human PR-B) in human subcutaneous adipose tissue. Using an enzymelinked immunosorbent assay, total PR could be quantitated. These studies substantiate that sex steroid receptors are present in human adipose tissue, thereby providing a direct route for regulation of adipose tissue by female sex steroids. (J Clin Endocrinol Metab 83: 509 –513, 1998) S Subjects and Methods EX steroids are postulated to influence adipose tissue regulation and distribution, because changes in fat distribution coincide with onset of ovarian production of estrogen and progesterone during puberty and with cessation of hormone production during menopause (1). Women typically demonstrate a gynoid or lower body fat distribution, whereas women with excessive androgens and glucocorticoids manifest an android or upper body fat distribution (2). Android adipose tissue distribution in women has been associated with an increased risk of diseases, including coronary artery disease (3), adult-onset diabetes mellitus (4), and endometrial cancer (5). Although clinical observations suggest that sex steroids influence adipose distribution, controversy exists as to whether sex steroids mediate these effects directly or indirectly in human adipose tissue. The dispute has been fueled by technical difficulty in detecting sex steroid hormone receptors in human adipose tissue (6, 7). Estrogen receptor (ER) messenger RNA (mRNA) and protein have been identified in human subcutaneous adipose tissue (8, 9). We have reported the presence of higher ER mRNA levels in the adipocyte fraction of adipose tissue compared with ER mRNA levels in the stromal-vascular portion (8). In this study, we demonstrated the presence of progesterone receptor (PR) mRNA and protein in subcutaneous adipose tissue of premenopausal women and examined the distribution of PR message in the two cell fractions. Subjects Subcutaneous adipose tissue was obtained from five premenopausal women undergoing elective adominoplasty with an average age of 43 6 4 yr and mean body mass index of 26.8 6 5.5 kg/m2. The included subjects were not taking exogenous hormones and were free of neoplastic, inflammatory, or autoimmune disease. All subjects signed an Institutional Review Committee-approved consent form (Greenville Hospital System) before surgery for use of tissue in research. Tissue acquisition and processing A portion of tissue was frozen immediately in liquid nitrogen for subsequent Western blot analyses and enzyme-linked immunosorbent assay (ELISA). Fresh adipose tissue was processed into adipocyte and stromal-vascular fractions as previously described (8). Both fractions were frozen in liquid nitrogen for Northern blot analyses. RNA isolation and Northern blot analysis Total RNA was isolated from whole abdominal adipose and myometrium using an RNeasy kit (Qiagen, Santa Clarita, CA) according to the manufacturer’s directions. Five micrograms of RNA was electrophoresed on a 1.2% denaturing agarose gel. The integrity of RNA was assessed by ethidium bromide staining of the gel. RNA was transferred to nylon by capillary blotting and fixed by ultraviolet cross-linking and vacuum baking. The filter was hybridized overnight at 65 C using a 32P random-prime labeled complementary DNA (cDNA) for human PR (10) added to prehybridization solution (7% SDS, 0.25 m sodium phosphate buffer, pH 7.2). Northern blots were washed at room temperature with 23 SSC and 0.1% SDS for 15 min followed by 23 SSC and 0.1% SDS for 15 min at 65 C and then exposed to film. For comparison of mRNA from the stromal-vascular and adipocyte cell fractions, total RNA was isolated as described above. Equal amounts of RNA from each fraction were denatured and applied to a nylon membrane housed in a slot-blot manifold. RNA was fixed and hybridized as described above. The autoradiograph was quantitated by scanning two-dimensional densitometry. The filter was then stripped and rehybridized with a 1.076-kilobase (kb) fragment of the mouse cytoplasmic b-actin gene (Ambion, Austin, TX). Following densitometry, PR values were normalized with b-actin values. The difference between the Received June 2, 1997. Revision received October 17, 1997. Accepted October 28, 1997. Address all correspondence and requests for reprints to: Susan N. O’Brien, Greenville Hospital System/Clemson University Biomedical Cooperative, Clemson University, 124 Long Hall, Clemson, South Carolina 29634-5112. 509 Downloaded from https://academic.oup.com/jcem/article/83/2/509/2865416 by guest on 18 June 2021 ABSTRACT Sex steroids are postulated to play a role in adipose tissue regulation and distribution, because the amount and location of adipose tissue changes during puberty and menopause. Because of the nature of adipose tissue, receptors for the female sex steroids have been difficult to demonstrate. To date, estrogen receptor messenger RNA and protein have been identified in human subcutaneous adipose tissue, but the presence of progesterone receptor (PR) has not been reported. In this study, we demonstrate PR message by Northern blot analysis in RNA isolated from the abdominal subcutaneous adipose 510 O’BRIEN ET AL. JCE & M • 1998 Vol 83 • No 2 ratio of PR to b-actin densitometry values for adipocytes compared with stromal-vascular cells in the five subjects were compared with a paired sample t test. Ammonium sulfate precipitation and Western blot analysis FIG. 1. Western blot analysis for hPR protein after ammonium sulfate precipitation. In myometrium (lane 1, 40 mg total protein), hPR-B isoform was identified at 116 kDa, whereas hPR-A isoform was primarily seen at 83 kDa, with a less intense band at 92 kDa. In adipose tissue (lane 2, 2 mg total protein), hPR-B isoform was identified at 116 kDa, whereas hPR-A isoform was primarily seen at 92 kDa. Control lane (lane 3, 2 mg adipose total protein) showed no specific bands when rabbit serum was used in place of primary antibody. A control lane with myometrial total protein also showing no specific binding is not shown. ELISA Frozen tissue was prepared in the same manner as for Western analyses. Protein was quantitated using Bradford analysis (BioRad). A 96-well plate was coated with a mixture of PR antibodies from NeoMarkers, PR antibodies 3, 4, 5, and 7 (Labvision, Freemont, CA) diluted to 1.5 mg/mL in 0.1 m sodium carbonate buffer (pH 9.3) at 100 mL/well. All subsequent volumes were 100 mL. The plate was incubated overnight at 4 C and then washed with PBS with 0.1% BSA and 0.05% Tween-20 (PBS-BSA-T) and blocked with 3% BSA in PBS. PR standard (Hormone Receptor Laboratory, Louisville, KY) at concentrations of 0, 3.75, 7.5, and 15 fmol/mg protein or 1 mg sample diluted with TEMMG were added to quadruplicate wells and incubated overnight at 4 C. After washing, a rabbit polyclonal PR antibody, C19, diluted to 0.5 mg/mL in PBS-BSA-T was added to triplicate wells. For a blank, one well included only PBS-BSA-T, without the primary C19 antibody. The plate was incubated overnight at 4 C. Biotinylated goat antirabbit IgG (Kirkegaard & Perry, Gaithersburg, MD) diluted 1:5000 in PBS-BSA-T was added to each well and incubated 1 h at room temperature. ImmunoPure Ultra-Sensitive ABC (Pierce, Rockford, IL) reagents were prepared according to the manufacturer’s directions, added to each well, and incubated 30 min at room temperature. Color developer (Horseradish Peroxidase Substrate Kit, BioRad) was added, and color development was measured after 1 h at 405 nm. Each sample was assayed in triplicate in three separate assays. The intraassay coefficient of variation (COV), calculated from values for the 7.5 fmol/mg protein standard was 4.5%. The interassay COV (11%) was calculated from the same standard values. Results Figure 1 shows the PR protein by Western blot analysis after ammonium sulfate precipitation in myometrium and abdominal adipose tissue. The human PR-B (hPR-B) isoform appeared at approximately 116 kilodaltons (kDa) and was identical in size to that found in the myometrium. The hPR-A isoform was located at approximately 92 kDa in adipose tissue, which was slightly greater than the predominant hPR-A band found at 83 kDa in the myometrium. A less intense third band in the myometrium was seen at 92 kDa FIG. 2. Northern blot analysis of hPR mRNA in adipose tissue (adip) and myometrium (uterus). Five micrograms total RNA was separated by denaturing gel electrophoresis, applied to nylon, and probed with a randomly primed labeled hPR cDNA. Bands of identical sizes were predominantly seen at 11.4, 6.1, 4.5, and 2.9 kb. corresponding to the hPR-A of the adipose tissue. Bands of these sizes were not detected in negative controls. In Fig. 2, Northern blot analysis of total RNA isolated from abdominal adipose tissue demonstrating the same size transcripts as seen in RNA isolated from myometrium is shown. The most prominent bands corresponded to sizes of 11.4, 6.1, 4.5, and 2.9 kb. Figure 3 shows the levels of PR protein in abdominal adipose tissue from five subjects and the level in myometrium from one subject as determined by ELISA. PR protein levels in adipose tissue varied from approximately 3– 8 fmol/mg total protein and were approximately 100-fold less than those found in myometrium. The insert shows a typical standard curve used for the PR ELISA. Downloaded from https://academic.oup.com/jcem/article/83/2/509/2865416 by guest on 18 June 2021 Frozen abdominal adipose tissue was ground with mortar and pestle in liquid nitrogen, added to TEMMG buffer (10 mm Tris-HCl, 1.5 mm EDTA, 12 mm monothioglycerol, 10 mm sodium molybdate, 10% glycerol) with protease inhibitors (10 mg/mL leupeptin, 1 mm phenylmethylsulfonyl fluoride, 10 mg/mL aprotinin, 1 mg/mL pepstatin), and homogenized on ice. As a positive control, uterine tissue was prepared in the same manner as adipose tissue. Following centrifugation for 20 min at 16,000 3 g, the layer between the floating lipid and pellet was removed. Ammonium sulfate was added to an aliquot of the cytosol to 30% saturation and rotated overnight at 4 C. The mixture was centrifuged at 3000 3 g for 30 min. Precipitates were resuspended in TEMMG plus SDS sample buffer and boiled for 2 min before loading. Samples were electrophoresed on a 7.5% polyacrylamide gel and transferred to polyvinylidene difluoride (Bio-Rad Labs., Hercules, CA). The membrane was incubated with rabbit antihuman PR (C19, Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4 C. As a second control, duplicate lanes were incubated with rabbit sera in place of the primary antibody. The membrane was washed and incubated with secondary antibody, peroxidase-conjugated goat antirabbit IgG (Cappel, ICN Pharmaceuticals, Costa Mesa, CA). Proteins were detected using an Enhanced Chemi Luminescense Western blotting detection system (Amersham, Buckinghamshire, England) according to the manufacturer’s directions. PR IN HUMAN ADIPOSE TISSUE 511 Figure 4 shows PR mRNA levels in adipocytes compared with stromal-vascular cells from abdominal adipose tissue. Northern blot analysis (Fig. 4A) for PR, performed with total RNA after denaturing gel electrophoresis, demonstrated greater PR mRNA levels in adipose stromal cells compared with adipocytes. Slot-blot analysis (Fig. 4B) was performed to semiquantitate the difference in PR mRNA levels between the two cell fractions in five subjects. Northern blot analysis of b-actin was used to control for differences in loading of total RNA. Densitometry values for the slots were expressed as the ratio of the value for the PR and the corresponding value for b-actin. Stromal-vascular cells contained approximately 2.5-fold greater levels of PR mRNA compared with adipocytes. Discussion A substantial amount of clinical evidence indicates that sex steroids may play a role in fat tissue regulation and distribution. Prepubertal boys and girls do not differ significantly in adipose tissue percentage or distribution. However, with puberty, females experience an increase in fat percentage as well as selective adipose tissue deposition in the breast and gluteo-femoral regions. This gynoid fat distribution commonly exists until menopause in normal nonobese women. Women with disorders of androgen excess often display an upper body or android type of fat distribution characterized by abdominal adipose deposition and a higher waist-to-hip ratio (2). During menopause, at which time ovarian production of estrogen and progesterone ceases, the female phenotype becomes more android (11, 12). Because upper body fat distribution is strongly correlated with pathophysiology (13), the understanding of sex steroid regulation of adipose tissue distribution may be important in ultimately preventing this process and decreasing disease risk. FIG. 4. A, Northern blot analysis comparison of PR mRNA levels in adipocytes vs. stromal-vascular cells. Five micrograms total RNA was separated by denaturing gel electrophoresis, applied to nylon, and probed with a randomly primed labeled hPR cDNA. Insert, 28S and 18S ribosomal RNA bands of ethidium bromide gel. B, Slot-blot analysis was used to semiquantitate PR mRNA levels in adipocytes (Adip) vs. stromal-vascular (SV) cells of samples from five subjects. Five micrograms total RNA was applied to nylon and probed with a randomly primed labeled hPR cDNA. The membrane was then stripped and reprobed with a labeled b-actin cDNA. Slot-blots were quantitated with laser densitometry, and results expressed as ratio of PR to b-actin. For each subject, samples were run in triplicate, with SD bars showing variance. A paired t test for all patients showed a significant difference between SV and Adip, (P # 0.001). A representative slotblot is shown. Detection of sex steroid receptors has been technically difficult in human adipose tissue because of the inherent properties of the tissue. In past studies, ERs could not be demonstrated with routine binding assays because of the high nonspecific binding of estrogens stored in adipose tissue (14). With the development of more sensitive assays, ER mRNA (by PCR and Northern blot analysis) and ER protein (8, 9) have been demonstrated in human adipose tissue. To date, there have been no reports of the presence of PR mRNA or protein (7, 15) in human adipose tissue, although specific progestin binding has been demonstrated in the adipose tissue of rat (16) and sheep (17). In addition to traditional target tissues, PR has been identified in such nonclassical human tissues as osteoclasts (18), the peripheral veins (19), and prostate stromal cells (20). In this study, we were able to demonstrate PR mRNA by Downloaded from https://academic.oup.com/jcem/article/83/2/509/2865416 by guest on 18 June 2021 FIG. 3. Quantitation of PR protein levels by ELISA in adipose tissue of five subjects and from one myometrium sample. SD bars represent variation for each subject within three separate assays. Insert, A typical standard curve using PR standards as described in Subjects and Methods. 512 JCE & M • 1998 Vol 83 • No 2 O’BRIEN ET AL. to investigate the target genes for these sex steroids in this tissue. Ultimately, these studies will help clarify the role of sex steroids in the regulation and distribution of adipose tissue. Acknowledgments We thank Dr. Geoffrey Greene for his kind gift of human PR cDNA. We gratefully acknowledge the technical assistance of Stephanie Weathers (Greenville Hospital System/Clemson University Biomedical Cooperative) and Dawn Blackhurst (Greenville Hospital System, Department of Medical Research). References 1. Marin P, Bjorntorp P. 1993 Endocrine-metabolic patterns and adipose tissue distribution. Horm Res. 39:81– 85. 2. Kirschner MA, Samojlik E. 1991 Sex hormone metabolism in upper and lower body obesity. Int J Obes. 15:101–108. 3. Lapidus L, Bengtsson C, Larsson, et al. 1984 Distribution of adipose tissue and the risk of cardiovascular disease and death: a 12-yr follow-up of participants in the population study of women in Gothenburg, Sweden. Br Med J. 289:1257–1261. 4. Kisselbal AH, Peiris ANA. 1989 Biology of regional body fat distribution: relationship to non-insulin-independent diabetes mellitus. Diabetes Metab Rev. 5:83–109. 5. Schapira DV, Kumar NB, Lyman GH, et al. 1991 Upper body fat distribution and endometrial cancer risk. JAMA. 266:1808 –1812. 6. Rebuffe-Scrive M, Bronnegard M, Nilsson A, et al. 1990 Steroid hormone receptors in human adipose tissue. J Clin Endocrinol Metab. 71:1215–1219. 7. Bronnegard M, Ottoson S, Boos J, et al. 1994 Lack of evidence for estrogen and progesterone receptors in human adipose tissue. J Steroid Biochem Mol Biol. 51:275–281. 8. Price TM, O’Brien SN. 1993 Determination of estrogen receptor mRNA and cytochrome P450 aromatase mRNA levels in adipocytes and adipose stromal cells by competitive polymerase chain reaction amplification. J Clin Endocrinol Metab. 77:1041–1045. 9. Mitzutani T, Nishikaw Y, Adachi H, et al. 1994 Identification of estrogen receptor in human adipose tissue and adipocytes. J Clin Endocrinol Metab. 78:950 –954. 10. Misrahi M, Atger M, d’Auriol, et al. 1987 Complete amino acid sequence of human progesterone receptor deduced from cloned cDNA. Bioch Biophys Res Comm. 143:740 –748. 11. Wing RR, Matthews KA, Kuller LH, et al. 1991 Weight gain at the time of menopause. Arch Intern Med. 151:97–102. 12. Pasquali R, Casimirri F, Labate Amm, et al. 1994 Body weight, fat distribution, and the menopausal status in women. Int J Obes. 18:614 – 621. 13. Baumgartner RN, Heymsfield SB, Roche AF. 1995 Human body composition and the epidemiology of chronic disease. Obesity Res. 3:73–95. 14. Watson GH, Manes JL, Mayes JS, McCann JP. 1993 Biochemical and immunological characterization of oestrogen receptor in the cytosolic fraction of gluteal, omental, and perirenal adipose tissues from sheep. J Endocrinol. 139:107–115. 15. Pedersen SB, Fuglsig S, Sjogren P, Richelsen B. 1996 Identification of steroid receptors in human adipose tissue. Eur J Clin Inves. 26:1051–1056. 16. Gray JM, Wade GN. 1979 Cytoplasmic progestin binding in rat adipose tissues. Endocrinology. 104:1377–1382. 17. Mayes JS, McCann JP, Ownbey TC, Watson GH. 1996 Regional differences and up-regulation of progesterone receptors in adipose tissues from oestrogentreated sheep. J Endocrinol. 148:19 –25. 18. Boivin G, Terrier-Anthoine C, Morel G. 1994 Ultrastructural localization of endogenous hormones and receptors in bone tissue: an imunocytological approach in frozen samples. Micron. 25:15–27. 19. Bergqvist A, Bergqvist D, Gerno M. 1993 Estrogen and progesterone receptors in vessel walls: biochemical and immunological assays. Acta Obstet Gynecol Scand. 72:10 –16. 20. Mobbs BG, Liu T. 1990 Immunohistochemical localization of progesterone receptors in benign and malignant human prostate. Prostate. 16:245–251. 21. McDonnell DP. 1995 Unraveling the human progesterone receptor signal and transduction pathway. Trends Endocrinol Metab. 6:133–138. 22. Wei LL, Krett NL, Francis MD, et al. 1988 Multiple progesterone receptor ribonucleic acids and their autoregulation by progestin agonists and antagonists in breast cancer cells. Mol Endocrinol. 2:62–72. 23. Gromeyer H, Meyer ME, Bocquel MT, et al. 1991 Progesterone receptors: isoforms and antihormone activity. J Steroid Biochem Mol Biol. 40:271–278. 24. Wen DX, Xu Y, Mais DE, et al. 1994 The A and B isoforms of the human progesterone receptor operate through distinct signalling pathways within target cells. Mol Cell Biol. 14:8356 – 8364. Downloaded from https://academic.oup.com/jcem/article/83/2/509/2865416 by guest on 18 June 2021 Northern blot analysis and PR protein by Western blot analysis in the abdominal subcutaneous adipose tissue of premenopausal women. Discrete transcripts of sizes identical to those present in uterine tissue can be detected in Northern blot analyses of total RNA isolated from human adipose tissue, suggesting that regulation of transcription for PR is as complicated in adipose tissue as in reproductive tissues (21). These multiple transcripts are obtained from two promoters, yielding hPR-A- and hPR-B- specific transcripts (22, 23), the identities of which have not been determined. The physiological response to progesterone is conveyed by two discrete forms of the PR (hPR-A and hPR-B) in target tissues (24). The B isoform is a 933-amino acid protein that acts primarily as an activator of transcription. hPR-A (768 amino acids) is an N-terminally truncated form of the larger B protein lacking the first 164 amino acids and functions as a repressor of hPR-B activity (25). In addition, hPR-A can repress transcriptional activation of glucocorticoid, mineralocorticoid, and ERs, suggesting that PR may play a pivotal role in the regulation of other steroid hormones (26, 27). Both the A and B isoforms of the PR appear to be present in human adipose tissue. The B isoform was of similar size to that of uterine tissue at approximately 116 kDa. The A isoform in adipose tissue was slightly greater in size (92 kDa) than the predominant A isoform in uterine tissue used in this study as a positive control (83 kDa). This may be because of differences in the degree of phosphorylation of the protein, which has been described in other tissues (28). The ratio of the two PR isoforms has been shown to vary in chick oviduct tissue according to estrogen status (29). Whether the ratio of the isoforms will vary in adipose tissue with changes in endogenous estrogen levels remains to be investigated. Further analysis of the PR involved quantitation of the protein levels by a multiple antibody ELISA. The values of the PR in human abdominal adipose tissue of premenopausal women are similar to those reported by Watson et al. (14) in sheep and are much lower than the uterine control, as expected. Differences in the cellular content of PR in human adipose tissue are also evident. In this study, we demonstrated higher levels of PR mRNA in the stromal-vascular cells compared with adipocytes. Previous studies have shown ER mRNA levels to be higher in adipocytes compared with stromalvascular cells (8). In addition, cytochrome P450 aromatase, the enzyme responsible for conversion of C19 steroids into C18 estrogens, has higher activity and mRNA levels in stromal-vascular cells compared with adipocytes (8, 30). These observations suggest a differential type of regulation by sex steroids of the cells in human adipose tissue. Estrogen of ovarian origin or estrogen produced by aromatization in adipose stromal cells may transcriptionally regulate adipocytes. In contrast, progesterone of ovarian origin may primarily regulate adipose stromal cells. This selective distribution of ER and PR levels in different cells of a given tissue has also been reported with endometrium of the rabbit uterus (31), where PR levels are higher in endometrial stromal cells compared with glandular cells, but ER levels are higher in glandular cells. Now that the presence of the ER and PR in human adipose tissue has been established, further studies will be possible PR IN HUMAN ADIPOSE TISSUE 25. Vegetto E, Shabaz M, Wen DX, et al. 1993 Human progesterone receptor A forms is a cell- and promoter-specific repressor of human progesterone receptor B function. Mol Endocrinol. 7:1244 –1255. 26. McDonnell DP, Shahbbaz MM, Vegeto E, Goldman ME. 1994 The human progesterone receptor A-form functions as a transcriptional modulator of mineralocorticoid receptor transcriptional activity. J Steroid Biochem Mol Biol. 48:425– 432. 27. McDonnell DP, Goldman ME. 1994 RU486 exerts antiestrogenic activities through a novel progesterone receptor A form-mediated mechanism. J Biol Chem. 269:11945–11949. 28. Clarke CL, Feil PD, Satyaswaroop PG. 1987 Progesterone regulation by 17b- 513 estradiol in human endometrial carcinoma grown in nude mice. Endocrinology. 121:1642–1648. 29. Syvala H, Vienonen A, Ylikomi T, et al. 1997 Expression of the chicken progesterone receptor forms A and B is differentially regulated by estrogen in vivo. Biochem Biophys Res Commun. 231:573–576. 30. Cleland WH, Mendelson CR, Simpson ER. 1983 Aromatase activity of membrane fractions of human adipose tissue, stromal cells, and adipocytes. J Endocrinol. 113:2155–2157. 31. Zaino RJ, Clarke CL, Feil PD, Satyaswaroop PG. 1989 Differential distribution of estrogen and progesterone receptors in rabbit uterus detected by dual immunofluorescence. Endocrinology. 125:2728 –2734. The National Institute on Aging announces the annual Summer Institute on Aging Research, a week-long workshop for new investigators, focused on current issues, research methodologies, and funding opportunities. The program will also include consultations on the development of research interests. The 1998 Summer Institute will be held July 11–17 in Airlie, Virginia. Support is available for travel and living expenses. Applications are due March 13, 1998. To increase the diversity of participants, minority investigators are encouraged to apply. For more information write to: Zita E. Givens, National Institute on Aging, National Institutes of Health, Building 31, Room 5C-35, 31 Center Drive MSC-2292, Bethesda, Maryland 20892-2292; or Telephone: 301496-0765; Fax: 301-496-2525; E-Mail: [email protected]; Web Site: http://www.nih/gov/nia. Downloaded from https://academic.oup.com/jcem/article/83/2/509/2865416 by guest on 18 June 2021 Summer Institute on Aging Research Airlie, Virginia July 11–17, 1998