Methods and Results of Aromatization Studies in Vivo1

Home , Aromatization

[CANCER RESEARCH (SUPPL.) 42, 3307S-3311s, August 1982] 0008-5472/82/0042-OOOOS02.00 Methods and Results of Aromatization Studies in Vivo1

Christopher Longcope

Department of Obstetrics/Gynecology and Medicine, University of Massachusetts Medical School, Worcester, Massachusetts 01605

Abstract peripheral extraglandular tissues, aromatization by these tis sues was not established until the work of MacDonald ef al. In order to characterize aromatization in vivo, an androgen (18) and Longcope ef al. (15). These workers administered and an estrogen, labeled with different radioactive isotopes, radiolabeled androgens to normal subjects and after extensive are administered simultaneously, either as a pulse alone or as purification procedures found radiolabeled estrogens in the a pulse followed by a constant infusion of the same radiolabeled urine and blood. Studies since then have been directed at steroids. Subsequently, blood samples are obtained at varying time intervals and analyzed for the 3H/14C ratio in nonme- determining the importance of peripheral aromatization as a source of estrogens, the tissues involved in the reaction, dis tabolized unconjugated estrogen, and/or all urine is collected for 96 hr and analyzed for the 3H/14C ratio in an estrogen ease states which alter the rate of aromatization, and the controlling mechanisms of the reaction. These studies have conjugate. The fraction of androgen (An) entering the blood been carried out using analysis in either blood or urine, and I which is converted to and measured as a specific estrogen (Es) in the blood ((P]BBEs)is calculated from will review briefly the 2 techniques involved and then present and discuss the results. (3H-Es/14C-Es)B (3H-An/14C-Es),â€ž, Methods where (3H-Es/14C-Es)B is the ratio of radioactivities in the Analysis in Blood (2, 6, 15, 16). The subjects were given an i.v. specific estrogen in blood; (3H-An/14C-Es)lnj is the ratio of priming dose of an 3H-labeled androgen and a 14C-labeled estrogen in 0.9% NaCI solution followed by a constant i.v. infusion for 2 to 4 hr of radioactivities administered. The fraction of androgen admin the same radiolabeled steroids. During the course of the infusion, blood istered which is converted to a specific estrogen in the body samples were obtained from a vein in the arm opposite to the infusion. and measured as a specific estrogen conjugate in the urine ([P]BMES)iscalculated from In certain experiments, blood samples were obtained from multiple veins and an artery in order to establish tissue aromatization rates. (3H-Es/14C-Es)M The samples were analyzed for radioactivity as specific androgens and estrogens by standard methods involving solvent extraction fol (3H-An/'4C-Es>n, lowed by extensive purification procedures consisting of multiple chro- where (3H-An/14C-Es)M is the ratio of radioactivities in the matographies and derivative formation until radiochemical purity was achieved. The radioactivity in the samples was then measured by liquid specific estrogen conjugate. Irrespective of the method used, scintillation spectrometry, and the results were corrected for losses the aromatization of androstenedione is greater than the aro through the procedure. matization of testosterone; aromatization in normal men is The amount of radioactivity administered as 3H-labeled androgen greater than that in normal young women; aromatization in which enters the blood as the specific estrogen during the infusion is postmenopausal women is greater than that in premenopausal calculated as women. 3H-Es x Aromatization occurs in many tissues but especially in adi pose tissue and muscle, and to a lesser extent in brain, kidney, where 3H-Es is the concentration of 3H in the specific estrogen and and skin. The liver appears to be a relatively minor site for MCREs is the metabolic clearance rate of the specific estrogen. The aromatization in normal subjects. fraction of 3H-labeled androgen administered which is aromatized to In certain diseases, e.g., hepatic cirrhosis and obesity, aro the specific estrogen [P]BB'ESiscalculated as matization rates are increased compared to those rates in normal individuals. rplBBE5 = 3H-Es x In men and postmenopausal women, peripheral aromatiza where rAn is the rate of infusion of 3H-labeled androgen. This can also tion of androgen is a major source of circulating estrogens. The peripheral aromatization of an androgen does not appear be expressed as to be influenced by luteinizing hormone, follicle-stimulating (3H-Es/'4C-Es)B hormone or adrenocorticotropic hormone or by substrate or (3H-An/'4C-Es),â€ž, product concentration. Blood flow through peripheral tissues may play a major role in control of this physiologically important where (3H-Es/14C-Es)B is the 3H/14C ratio of radioactivity administered reaction. as the specific estrogen in the blood and (3H-An/14C-Es)ini is the 3H/ 14Cratio of administered radioactivity, both androgen and estrogen. This method gives the fractional rate of aromatization of androgen Although there were indications from the early work of Leach to estrogen where the estrogen must enter the blood pool of circulating et al. (10) that androgens could be aromatized to estrogens by unmetabolized estrogen, i.e.. Chart 1, Step 2. Inherent in the technique 1 Presented at the Conference "Aromatase: New Perspectives for Breast is the establishment of radiochemical purity as noted and also the Cancer," December 6 to 9, 1981, Key Biscayne. Fla. This work was supported achievement of an isotopie steady state for the androgen and estrogen in part by USPHS Grants NICHHD 15443 and NICHHD 08610. radioactivity in the blood during the infusion.

AUGUST 1982 3307s

With this type of analysis, the production rates of androgens and from androgen in peripheral tissues. However, [p]EMEsis a measure of estrogens are determined at the same time as the fractional aromati Step 1 which represents all the estrogen formed from androgen in the zation rates are calculated. Therefore, the mass of androgen aroma tissues, whether it enters the blood (Step 2) or is conjugated to EConÂ¡ tized to estrogen and the importance of this aromatization can be (Step 3) before it enters the blood. Therefore, determined. Analysis in Urine (17, 18). The subjects received radiolabeled = [p]eSEs androgens and estrogens either as a pulse or as a priming dose, If Step 3 is nonexistent, then [P]BMEÂ°willbe equal to[p]ÃŸBEs.However, followed by a constant infusion of the same radiolabeled steroids. if Step 3 is present, then [P]BBESwill be smaller than [p]auiEs bV an Following the administration of radioactivity, all urine was collected for 3 to 5 days and pooled. An aliquot equal to 50 to 100% of the urine amount equal to Step 3. An additional cause for a discrepancy between [P]BMESand [p]EBEs was extracted with an organic solvent to remove any free steroid, and the urine was then subjected to /3-glucuronidase hydrolysis. After would occur if an isotopie steady state were not reached for estrogen hydrolysis, the urine was extracted with organic solvent, and the in the blood during the infusion of androgen. In such a situation, unconjugated steroids were purified by multiple chromatographies and M&E' > [p]ESE5 derivative formation. Because the amounts of radioactivity involved are larger than in the blood specimens, most procedures have included Since urine is collected until essentially all radioactivity as estrogen is crystallization to constant specific activity for final purification. excreted, [P]BMESwill be independent of the isotopie steady state in The fractional rate of aromatization was calculated as blood for estrogen. _ (3H-Es/'4C-Es)M With this type of analysis, blood samples must still be analyzed to (3H-An/"C-Es),n, provide the data necessary for calculating blood production rates of androgens. These production rates will be needed to determine the where (3H-Es/MC-Es)M was the ratio of radioactivity as 3H to 14Cin the mass of androgen converted to estrogen using the [P]BMESvalues. urinary estrogen conjugate. As shown in Chart 1, [P]BBESis a measure of Step 2 which represents Results all the unconjugated estrogen entering the blood after its aromatization Overall Aromatization. As shown in Table 1, the mean values for aromatization in men are greater than are the respective mean values in women, and in either sex the mean value for TISSUE [pies51 is greater than that for [pJBiP- It should be noted that these studies were all done in subjects who were within 20% of their ideal weight. These values are not different from the [P]BM' and [p]eM2 values reported by MacDonald ef al. (1 8). The similarity of the values in these normal subjects indicates that the previously mentioned step (Chart 1, Step 3) is not of major importance. In addition, Steps 1 and 2 appear to be relatively rapid, and an isotopie steady state was obtained during the infusion for the radiolabeled estrogen in the blood. The sex difference in [pleif1 values can be removed if the [P]BBEIvalues are normalized for body weight (Table 1). The mean values of [piel2, however, remain significantly different for men and women even if normalized for body weight. There fore, we conclude that the sex hormone-binding globulin of testosterone may, in part, play a role in the difference in [p]eiP values between men and women which cannot be ex plained by differences in body weight and/or composition. BB An The actual mass of androgen aromatized to a specific estro URINE gen (Pen'Es)can be calculated as Chart 1. Peripheral aromatization as measured by analysis in blood and analysis in urine. Radiolabeled androgen (An) is infused into blood and trans [p]EKiE ported into tissues where it is aromatized to estrogen (Es) (Step 1). Estrogen may then enter the blood pool of estrogen (Step 2) or be transformed to estrogen conjugate (Es cony) (Step 3). Estrogen in blood is later conjugated, and all or estrogen conjugate is excreted in urine. Analysis in blood measures Step 2 while lAn.Es analysis in urine measures Step 1. [p] X PÃŠ

Table 1 IPÃŒBB'andMai' in normal men and women with and without normalization tor weight All measurements were carried out using analyses in blood.

Â±0.0005s Men Â±0.000007 Â±0.0026 Â±0.000004 Womenp"n15 29[PAP0.00340.0018 Â±0.0001<0.01[pEIVkg0.0000470.000030 Â±0.000003<0.02n14 34[PAT0.01770.0107 Â±0.0008<0.01[pjaf'/kg0.0000240.000018 Â±0.000002>0.1

Mean Â±S.E. ' p values for difference between men and women.

3308s CANCER RESEARCH VOL. 42

Methods and Results of Aromatization Studies in Vivo

where PB" is the blood production rate of the androgen. The than [plÃ¢iF where the latter is measured using the standard relative importance of PenEs to the production rate of the short infusion time (5). As the time of infusion is increased, estrogen Pls can be calculated as then [P]B!' gradually rises to roughly equal [p]e&' (4). This is not due to an increase in Step 3 (Chart 1) but rather is due to pAn.E./p|. the slow entry into the blood, via Step 2, of radiolabeled estrogen formed via Step 1 (4). As shown in Table 3, in Although the fractional aromatization rates themselves are individuals who are of normal weight, small, because of the magnitude of the androgen production compared to estrogen production rates, the amount of andro gen aromatized will contribute substantially to the amount of estrogen produced. In men, this is the source of 60 to 75% of but, in subjects who are overweight, r-lAn.Es ^ r lAn.Es the estrogen produced each day. In normal young women, IPJUB < [PlBM peripheral aromatization of androgens is not a source of estra dici and is only slightly more important as a source of estrone. Therefore, in measuring [p]A E' or [p]T-Ei in obese subjects, the In other situations, however, peripheral aromatization becomes urinary method should be used. far more important. Since adipose tissue is responsible for a considerable portion Aromatization in Specific Tissues. Much of the work on of the overall [p]An'Esmeasured in blood or urine, it is not tissue aromatization has been carried out in vitro, in part due surprising that, as the adipose tissue mass increases, there is to the difficulty in sampling the venous drainage of multiple an increase in [p]*"^8. This increase is directly correlated to the tissue in vivo, especially in humans. However, by catheteriza- increase in weight (5, 17, 20). Peripheral aromatization in tion of specific forearm veins draining muscle or draining obese men, as in normal men, is the major source of circulating adipose tissue, we have carried out studies on the aromatiza estrogen and also becomes the major source of estrogens in tion of these tissues in normal men (16). We found aromatiza many obese women (9,17). From studies in vitro, Ackerman et tion to occur in both muscle and adipose tissue in the forearm al. (1 ) have shown that the stromal cell and not the adipocyte at low but essentially equal fractional rates. We estimated the is the specific cell type containing the aromatase system. contribution of adipose and muscle tissue to the overall [P]BBEI In obese individuals who are studied before and immediately value using the overall blood flow to these tissues reported in after losing weight, the (/>]Â£Â£'remainsthe same (21). Whether the literature (11, 23). In this manner, we calculated that over this is related to a persistence of the stromal cells, an increase 50% of the aromatization occurred in adipose tissue and mus in blood flow (11), or other reasons remains to be determined. cle. It should be realized that these are rough estimates since Cirrhosis of the Liver. Men with cirrhosis of the liver fre they depend on homogeneity of tissue and calculated blood quently show signs of hyperestrogenism (3). Initially, this was flows. thought to be due to a decrease in estrogen metabolism; There are few data quantitating aromatization across other however, recent studies have shown that cirrhotic males have tissues in vivo. Preliminary studies in rhesus monkeys indicate that kidney, skin, and brain all contribute to the overall aro Table 2 matization, but the contribution of the liver appears to be I pirn 2ar*d IPÃŒBB''" young reproductive-age and older postmenopausal women minimal.2 Â±1.0661 Conditions Associated with Altered Aromatization Rates. YoungwomenPostmenopausalwomenYoung Â±0.00010.0065 The first situation in which aromatization rates were noted to .3 Â±1.429.2 Â±0.0016IplalF0.01 be increased was in the postmenopausal woman. Studies by Grodin ef al. (8) and our own studies (13) (Table 2) indicated a marked increase in the aromatization of both androstenedi- womenPostmenopausalwomenn29323445Age26.9Â±1.265.6 1 Â±0.00010.022 one and testosterone in postmenopausal women as compared Â±1.5[PEP0.0018 Â±0.002p'<0.01<0.01 to premenopausal women. This increase appears to occur around the time of the menopause but does not appear to be p value for difference between young and postmenopausal women. age dependent. If one takes women of all ages, there is a " Mean Â±S.E. correlation between the [P]BB' value and age. However, if one Table 3 takes the premenopausal and postmenopausal groups sepa Ratio of[pleÂ£E'to [p]e!iE< in subjects ol differing weights rately, then for neither group is there a significant correlation between the [P]BB' value and age. The reason for the increase Mean wt (% ofideal)1 in peripheral aromatization in postmenopausal women remains 10 Â± 36 Â±0.13 uncertain. However, in this group, the major if not the only 0.68>ÃŠ2iEV|p]Ã¢BEt.223 Â±23EM/BE"1.05 4.34 Â± source for circulating estrogens is the peripheral aromatization Mean Â±S.E. of androgens (8, 13, 14). Obesity is associated with a marked increase in aromatiza Table 4 tion (20), and morbidly obese individuals were noted to have Peripheral aromatization in normal and cirrhotic men [P]BMESvalues which ranged up to 10 times that in individuals 0.0482 Â±0.0132a of normal weight (5, 17). Of theoretical and practical interest is Cirrhotics 0.0067 Â±0.0021 the fact that, in obese subjects, [P]BM' is considerably greater Normals 0.0177 Â±0.0026 0.0034 Â±0.0005 p" <0.01 >0.10 Mean Â±S.E. 2 C. Longcope, R. Billiar, and B. Little, unpublished observations. ' p values for difference between cirrhotics and normals.

AUGUST 1982 3309s

Table 5 Effects of human chorionic gonadotropin, follicle-stimulating hormone, estrogen, and testosterone administration and castration on the[piaec' andlpfe/' in male Rhesus monkeys

0.001aÂ± ControlHuman Â±0.0030.0021 gonadotropinFollicle-stimulatingchorionic days500000 units each day for 5 0.0035Â± hormoneHuman days1units each day for 5 0.004Â± +follicle-stimulatingchorionic gonadotropin units1000 + 500 0.0032Â± hormoneEstrogenCastrateTestosteroneEstrogenTreatment1 days300 pg each day for 8 0.008Â± wk5 0.001IpfeP0.0024

days1mg each day for 7 Â±0.00050.0009 0.0005" 00 fig twice a wk for 2 wk0.0150.00950.0160.0130.0150.008\pJEtTÂ± Â± Mean Â±S.E. an increase in the peripheral aromatization of androgens (7). should be noted that estrogen administration to male monkeys As our data indicate (Table 4), this increase is more marked for did result in a decline in the fractional aromatization, but this the aromatization of androstenedione than for testosterone. It decline was not significant due to the wide range of individual is probable that the increase in sex hormone-binding globulin values and the small number of animals involved. However, in cirrhotics (7) may inhibit the entry of testosterone into the there appears to be no indication that gonadotropins or other tissues and thus render it less available as a substrate for polypeptide hormones influence peripheral aromatization di aromatization, but androstenedione, which is not bound to the rectly. This lack of influence is interesting in view of the marked sex hormone-binding globulin, would all be available to the increases in gonadotropins and the fractional aromatization tissues. The failure to find a high degree of aromatization in rates which occur at the menopause. normal hepatic tissues2 raises an interesting point as to whether Thus, aromatization appears to occur in many tissues, but the increase in aromatization in hepatic cirrhosis in individuals primarily in adipose tissue and muscle, and can be a major is due to an increase in aromatization in the liver itself or to an source of circulating estrogens. Increases in the aromatization increase in aromatization in peripheral, nonglandular tissue. rate can be present in various diseases and can result in signs Since blood flow is increased in peripheral tissues in many of hyperestrogenism in these individuals. Mechanisms under instances of cirrhosis (24), it is probable that this increased lying control of the peripheral aromatization are still obscure, blood flow results in more androgen substrate available to the but blood flow and availability of androgen substrate to specific tissues and therefore greater aromatization and that the liver tissues may play a critical role in this control. contributes little to the increase in aromatization. Cancer. The fractional rate of aromatization has been ex References amined in various endocrine-related cancers such as endometrial carcinoma and breast carcinoma. Although initial re 1. Ackerman, G. E., Smith, M. E., Mendelson. C. R., MacDonald, P. C., and Simpson, E. R. Aromatization of androstenedione by human adipose tissue ports suggested an increased aromatization in women with stremai cells in monolayer culture. J. Clin. Endocrinol. Metab., 53. 412- endometrial cancer compared to normal women, MacDonald 417. 1981. 2. Baird, D. T., Horton, R., Longcope. C.. and Tait. J. F. Steroid dynamics et al. (17) and Rizkallah ef al. (20) have shown that, in women under steady-state conditions. Recent Prog. Horm. Res., 25. 611-664. with endometrial cancer, when matched for weight and meno- 1969. 3. Baker, H. W. G., Burger, H. G., de Kretser, D. M., Dulmanis, A., Hudson, B., pausal status, the aromatization rates are similar. Studies in O'Connor, S., Paulsen, C. A., Purcell, N., Rennie, G. C., Seah, C. S., Taft, breast cancer have given equivalent results (19, 21 ), and there H. P., and Wang, C. A study of the endocrine manifestations of hepatic appears to be no direct increase in aromatization related to cirrhosis. Q. J. Med.. )77: 145-178, 1976. breast cancer. Although breast cancer tissue appears to con 4. Edman, C. D., and MacDonald, P. C. Slow entry into blood of estrone produced in extraglandular site(s) in obesity and endometrial neoplasia. tain the aromatase (12) system, the contribution of the cancer Gynecol. Invest. 5. 27, 1974. to the overall aromatization rate would appear too low to alter 5. Edman, C. D., and MacDonald. P. C. Effect of obesity on conversion of plasma androstenedione to estrone in ovulatory and anovulatory young that rate. women. Am. J. Obstet. Gynecol., 130: 456-461, 1978. Miscellaneous Conditions. Initial studies by Southren et al. 6. Franz, C., and Longcope, C. Androgen and estrogen metabolism in male (22) in subjects with hyperthyroidism suggested that there was rhesus monkeys. Endocrinology, 705. 869-874, 1979. an increase in peripheral aromatization in this disorder. How 7. Gordon, G. G., Olivo, J., RafÃ¼,F., and Southren, A. L. Conversion of androgens to estrogens in cirrhosis of the liver. J. Clin. Endocrinol. Metab., ever, in recent studies,3 we have not been able to confirm this 40: 1018-1026, 1975. finding by measurements in either blood or urine. Further work 8. Grodin, J. M., Siiteri, P. K.. and MacDonald. P. C. Source of estrogen production in postmenopausal women. J. Clin. Endocrinol. Metab., 36. 207- may be necessary to clarify this point. Rizkallah ef al. (20) reported an increase in [p]e^' in a small group of postoperative 214, 1973. 9. Kirschner, M. A., Ertel, N., and Schneider, G. Obesity, hormones, and patients. The reason for this is uncertain. cancer. Cancer Res., 41: 3711-3717, 1981. 10. Leach, R. B., Maddock, W. O., Tokuyama, I.. Paulsen, C. A., and Nelson, W. Control Mechanisms. Since aromatization by peripheral tis O. Clinical studies of testicular hormone production. Recent Prog. Horm. sues is increased in a number of disparate conditions, we were Res., 12:377. 1956. 11. Lesser, G. T., and Deutsch, S. Measurement of adipose tissue blood flow interested in factors which might control the peripheral aro and perfusion in man by uptake of 85Kr. J. Appi. Physiol.. 23. 621-630, matization. Accordingly, we tested, in rhesus monkeys, a num 1967. ber of hormones and, as noted in Table 5, we were unable to 12. Li, K.. and Adams, J. B. Aromatization of testosterone and oestrogen receptor levels in human breast cancer. J. Steroid Biochem., 14: 269-272, influence the aromatization to any significant degree (6). It 1981. 13. Longcope, C. Metabolic clearance and blood production rates of estrogens 3 E. C. Ridgway. F. Maloof, and C. Longcope. unpublished observations. in postmenopausal women. Am. J. Obstet. Gynecol., 111: 778-781, 1971.

3310s CANCER RESEARCH VOL. 42

14. Longcope, C. The significance of steroid production by peripheral tissue. In: 19. Poortman, J., Thijssen, J. H. H., and Schwarz, F. Androgen production and E. R. Scholler (ed.), Endocrinology of the Ovary, pp. 23-35. Paris: Sepe, conversion to estrogens in normal postmenopausal women and in selected 1976. breast cancer patients. J. Clin. Endocrinol. Metab., 37. 101-109, 1973. 15. Longcope, C., Kato, T., and Morton. R. Conversion of blood androgens to 20. Rizkallah, T. H., Tovell, H. M. M. and Kelly, W. G. Production of estrone and estrogens in normal adult men and women. J. Clin. Invest., 48: 2191 -2201, fractional conversion of circulating androstenedione to estrone in women 1969. with endometrial carcinoma. J. Clin. Endocrinol. Metab., 40: 1045-1056, 16. Longcope, C., Pratt, J. H., Schneider, S. H.. and Fineberg, S. E. Aromati- 1975. zation of androgens by muscle and adipose tissue in vivo. J. Clin. Endocrinol. 21. Siiteri, P. K., Williams, J. E.. and Takaki. N. K. Steroid abnormalities in Metab., 46: 146-152. 1978. endometrial and breast carcinoma: a unifying hypothesis. J. Steroid Bio- 17. MacDonald, P. C., Edman, C. D., Hemsell, 0. L., Porter, J. C., and Siiteri, P. chem., 7: 897-903. 1976. K. Effect of obesity on conversion of plasma androstenedione to estrone in 22. Southren, A. L., Olivo, J., Gordon, G. G., Vittek. J., Brener. J.. and Rafii, F. postmenopausal women with and without endometrial cancer. Am. J. Obstet. The conversion of androgens to estrogens in hyperthyroidism. J. Clin. Gynecol.. 730. 448-455, 1978. Endocrinol. Metab., 38: 207-214, 1974. 18. MacDonald, P. C.. Rombaut, R. P., and Siiteri, P. K. Plasma precursors of 23. Wade, O. L., and Bishop, J. M. Cardiac Output and Regional Blood Flow, estrogen. I. Extent of conversion of plasma A'-androstenedione to estrone pp. 86-94 Oxford, England: Blackwell Scientific Publishers, 1962. in normal males and nonpregnant normal, castrate and adrenalectomized 24. Wade, O. L.. and Bishop, J. M. Cardiac Output and Regional Blood Flow, p. females. J. Clin. Endocrinol. Metab., 27: 1103-1111, 1967. 198. Oxford, England: Blackwell Scientific Publishers, 1962.

AUGUST 1982 3311s

Christopher Longcope

Cancer Res 1982;42:3307s-3311s.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/42/8_Supplement/3307s

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/42/8_Supplement/3307s. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.