help button home button Endocrine Society Endocrine Reviews
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sasano, H.
Right arrow Articles by Harada, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sasano, H.
Right arrow Articles by Harada, N.
Endocrine Reviews 19 (5): 593-607
Copyright © 1998 by The Endocrine Society

Intratumoral Aromatase in Human Breast, Endometrial, and Ovarian Malignancies1

Hironobu Sasano and Nobuhiro Harada

Department of Pathology (H.S.), Tohoku University School of Medicine, Sendai 980-8575, Japan; and Division of Molecular Genetics (N.H.), Institute for Comprehensive Medical Science, Fujita Health University, Toyoake 470-1192, Japan


   Abstract
Top
 Abstract
I. Introduction
 II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
 VI. Summary
 References
 

I. Introduction
II. Aromatase Cytochrome P450 and in Situ Estrogen Production
III. Aromatase in Breast Cancer
A. Introduction
B. Localization
C. Genetic regulation
D. Clinical and pathological correlation
E. Male breast cancer
IV. Aromatase in Endometrial Cancer
A. Introduction
B. Localization
C. Genetic regulation
D. Clinical and pathological correlation
V. Aromatase in Ovarian Cancer
A. Introduction
B. Localization
C. Genetic regulation
D. Clinical and pathological correlation
VI. Summary


   I. Introduction
Top
 Abstract
 I. Introduction
II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
 VI. Summary
 References
 
BIOLOGICALLY active steroids are produced and secreted in the endocrine organs, such as ovary, testis, and adrenal cortex, are transported through circulation, and act on their target tissues in which their specific nuclear receptors are present. Therefore, various biological features of steroid hormone-dependent target tissues are influenced by serum or plasma concentrations of these hormones. In the great majority of human breast and endometrioid endometrial cancers and in some ovarian cancers, estrogens, especially 17ß-estradiol (E2), a biologically potent estrogen, have been shown to contribute greatly to the growth and development of these neoplasms; indeed, some of these cancers actually require estrogen for their continued growth (1). Therefore, these cancers are considered to be estrogen-dependent neoplasms.

In women, estradiol originates from different sources. In premenopausal women, the ovary is the main source of circulating estrogens. However, after menopause, estrogens are produced through conversion of androgens of both adrenal and ovarian origin (2). The conversion of androgens to estrone has been shown to occur principally in peripheral tissues, including skin (3), muscle (4), fat (4), and bone (5). This conversion is catalyzed by the aromatase enzyme complex. Most of the estrone formed by aromatase in these peripheral tissues is then converted into estrone sulfate by estrone sulfotransferase, which is also present in these peripheral tissues (6). Estrone sulfate may act as a reservoir of estrone formation through estrone sulfatase, i.e., estrone sulfatase plays important roles in regulating the in situ availability of estrone in various peripheral tissues (7, 8). Estrone is subsequently reduced to 17ß-estradiol by 17ß-hydroxysteroid dehydrogenase (HSD) type I, which is also widely distributed in various peripheral tissues (9, 10, 11). Therefore, each of these enzymes is considered to play an important role in peripheral conversion of serum androgens to 17ß-estradiol but aromatization of androgens, which catalyzes the initial reaction of these conversion pathways, is generally considered to be the rate-limiting, or the most important, step of the pathway. Increased peripheral conversion of androgens to estrogens may result in elevated serum levels of estrogens. Therefore, numerous studies have been performed to study the subtle differences of serum estrogen concentrations and metabolism, which are derived from ovarian granulosa cells (12, 13), or peripheral tissues, as listed above in patients with sex steroid-dependent neoplasms. In breast cancer, several epidemiological studies indicate that plasma estradiol, adrenal androgens, and testosterone levels are higher in women who develop neoplasms over a period of several years than in those who do not (14, 15, 16). In particular, Berrino et al. (16) reported that high serum testosterone levels precede breast cancer occurrence. However, results of other studies (17, 18) were not necessarily consistent with those above. There has been no consistent evidence of increased serum estrogen concentrations or other systemic estrogen abnormalities reported in women with epithelial-stromal ovarian cancer (19, 20, 21) and endometrioid endometrial cancer (20, 22).

Miller et al. (23) and Perel et al. (24) independently demonstrated that human breast and its neoplasms can produce 17ß-estradiol in vitro. In addition, Reed et al. (25) directly demonstrated in situ estrogen synthesis in normal breast and breast tumor using an isotopic infusion technique. There was controversy as to whether aromatase and other enzymes involved in in situ estrogen production of human breast cancer can provide sufficient amounts of estrogens to stimulate tumor growth or exert various biological effects when these findings were first reported (26). Thorsen et al. (27) and van Landeghem et al. (28) subsequently demonstrated that the tissue concentrations of 17ß-estradiol in specimens of breast cancers from postmenopausal subjects were more than 10-fold higher than those in plasma. It is true, however, that the presence of aromatase in breast cancer tissue does not mean that aromatase is present in sufficient quantity to be biologically relevant and that the presence of high estradiol concentration does not necessarily indicate that the estrogen is made in tissue, since the possibility of enhanced uptake from plasma cannot be completely ruled out. However, very recently, Yue et al. demonstrated that in situ synthesis of estrogen predominates over uptake from plasma as a mean of maintaining estradiol concentrations in breast tissues after menopause. This was based on studies using xenografts of human breast cancer growing in ovariectomized nude mice (29) that were capable of synthesizing estrogen in situ by transfection of the aromatase gene. In addition, it was reported that exemestane, one of the aromatase inhibitors, caused maximal suppression of plasma estradiol and estrone to a mean of 14.6 and 5.8% of pretreatment levels, respectively, without any fall in adrenal steroid levels (30). Bezwoda et al. (31) also reported that tumor aromatization in breast cancer tissue was a useful measurement in predicting response to aromatase inhibitors. These findings indicate the biological importance of elevated in situ estrogen concentrations as a result of intratumoral aromatization in human breast cancer.

In endometrial cancer, Tseng et al. (32) examined testosterone aromatization in the human endometrium and its disorders and detected estrogenic products in these tissues. Yamaki et al. (33) also showed that aromatase activity is significantly higher in neoplastic endometrium than in normal tissues. These researchers then concluded that human endometrial neoplasms can directly convert androgens to estrogens and that this increased aromatization activity can result in increased in situ estrogen concentrations in neoplastic endometria. Increased aromatase activity has also been demonstrated in common epithelial or epithelial-stromal ovarian neoplasms by several groups of investigations (34, 35, 36). The biological significance of in situ estrogen production still remains controversial with regard to development and biological behavior of breast cancer and other estrogen-dependent neoplasms (2, 37, 38, 39). However, an increasing number of studies have indicated that in patients with estrogen-dependent breast, endometrial, and ovarian cancers, especially in postmenopausal women, intratumoral estrogens derived from in situ aromatization could function as an autocrine growth and mitogenic factor and could impart a growth advantage to these cancer cells, regardless of serum concentration of estrogens. Labrie and colleagues (40, 41) elegantly described the local formation of active androgens such as dihydrotestosterone from the inactive adrenal precursors, dehydroepiandrosterone, dehydroepiandrosterone-sulfate, and/or androstenedione, in some tissues or cells in adenocarcinoma of the prostate where biosynthesis takes place without release into the extracellular space as "intracrine activity" (40, 41). Therefore, estrogen-dependent neoplasms such as breast, endometrial, and ovarian cancers in which in situ conversions from serum androgens to biologically active estrogens occur may also be considered as "intracrine" tissues.


   II. Aromatase Cytochrome P450 and in Situ Estrogen Production
Top
 Abstract
 I. Introduction
 II. Aromatase Cytochrome P450...
III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
 VI. Summary
 References
 
As an initial step toward understanding intratumoral estrogen production and metabolism in estrogen-dependent neoplasms, it is very important to study whether or not aromatase is overexpressed in breast, endometrial, and ovarian cancer tissues. The aromatase enzyme complex is composed of two polypeptides, aromatase cytochrome P450 (42) and a flavoprotein, NADPH-cytochrome P450 reductase (42, 43). Aromatase cytochrome P450, also called P450arom or aromatase, is a (microsomal) unique member of the cytochrome P450 superfamily and is responsible for binding the C19 steroid substrate and catalyzing the series of reactions leading to formation of the phenolic A ring characteristic of estrogens (42). Aromatase (cytochrome P450 aromatase or P450arom) is specifically involved in the conversion of androgen to estrogen, i.e., aromatization, whereas NADPH-cytochrome P450 reductase is an essentially ubiquitous protein in the endoplasmic reticulum of most cell types and is responsible for transferring reduced equivalents from NADPH to any microsomal forms of cytochrome P450 with which it comes into contact (42). Therefore, it is very important to examine expression of aromatase in breast, endometrial, and ovarian neoplasms to obtain a better understanding of in situ or intratumoral estrogen production.

In humans, mapping of isolated genetic clones and Southern analysis of total genomic DNA indicated that the aromatase cytochrome P450 gene exists as a single-copy gene, spanning at least 70 kb (44, 45). The aromatase cytochrome P450 gene is also larger than other members of the cytochrome P450 superfamily and consists of 10 exons (44, 45). However, it is difficult to explain the complex transcriptional regulation of the aromatase cytochrome P450 gene that is widely expressed in human tissues by a single gene and/or a single promoter. The aromatase cytochrome P450 gene is characterized by the fact that exon 1, encoding the only 5'-untranslated region, is separated from exon 2 by an intron of more than 35 kb (44, 45, 46). Means et al. (47) and Mahendroo et al. (48) first demonstrated that the aromatase mRNAs in the human ovary and adipose stromal cells were transcribed from 79 and 84 bp upstream, respectively, of the exon 2 identified in placenta. These findings suggest that aromatase cytochrome P450 gene in human ovary and adipose stromal cells utilizes a novel exon 1 containing the placental exon 2 instead of the placental exon 1 and that aromatase expression in these tissues is regulated by a new promoter or promoters different from human placenta. This switching of the promoter or alternative utilization of exons 1 can explain, in part, the complex tissue-specific regulation of the human aromatase cytochrome P450 gene. It is now considered that all of these 5'-terminal exons 1 are spliced into a common junction upstream of the translation start site and the sequence encoding the open reading frame, i.e., the expressed aromatase protein is subsequently identical regardless of splicing patterns (46, 47, 48). Since the reports of Simpson and associates (47, 48), several different splicing variants present in aromatase transcripts have been independently reported by the groups of Simpson (42) and Harada (46, 49, 50). Therefore, different nomenclature for these splicing variants has been used by these two groups, as summarized in Table 1Go. Major exons 1 of the aromatase cytochrome P450 gene used as a promoter in aromatase gene expression in the human are summarized as follows: exon 1a or I.1, mainly used in placenta; exon 1b or I.4, used in skin and adipose fibroblasts and fetal liver; exon 1c or I.3, used in ovary; and exon 1d or P II, used in ovary, prostate, or testis (42, 46, 51). This utilization of alternative exons 1 of the aromatase cytochrome P450 gene can be examined without much difficulty by RT-PCR of the RNA fractions using sense primer specific for exon 1a (I.1), 1b (I.4), 1c (I.3), and 1d (PII) and the fluorescent dye-labeled antisense primer specific for exon 3 (46, 52, 53, 54) in various clinical materials as demonstrated in Figs. 1Go and 2Go.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of different nomenclature of major splicing variants of human aromatase gene and their predominantly used human tissues

 


View larger version (17K):
[in this window]
[in a new window]
  		Figure 1. Principle of evaluation of alternative splicing of multiple exons 1 in aromatase gene. Alternative splicing was examined by RT-PCR of the RNA fraction using sense primers specific for exons 1b (I.4), 1c (I.3), and 1d (PII) and the fluorescent dye-labeled antisense primer specific for exon 3. Fluorescent PCR products were subsequently analyzed with a Gene Scanner (Perkin Elmer Co., Foster City, CA). The aromatase transcripts from exons 1b (I.4), 1c (I.3), and 1d (PII) yielded PCR products at positions corresponding to 327, 368, and 355 bp, respectively.

 


View larger version (28K):
[in this window]
[in a new window]
  		Figure 2. An example of typical data of RT-PCR study of utilization of alternative exon 1 in human breast cancer tissues. In the figure, 262, 293, 317, and 439 represent the peaks of the GeneScan 1000ROX of the internal size standards. In patient 1 (61 yr old), exon 1b or I.4 fibroblast types of alternative exons 1 of aromatase gene (exon 1) were predominantly used. In patient 2 (53 yr old), the major transcript was exon 1c or I.3 gonadal type of alternative exons 1 of aromatase gene (exon 1) with 1d or PII and 1b or I.4 as minor transcripts. Analysis of alternate exon 1 is rather semiquantitative, but different curves were adequately resolved judging from PCR analysis using two different types of fluorescent dye-labeled primers.

 
This alternative splicing of the aromatase cytochrome P450 gene is also considered to play important roles not only in tissue-specific expression of aromatase but also in alteration of aromatase expression through developmental or neoplastic process of the same tissue. Harada et al. demonstrated that a major transcript using exon 1b (I.4) and a minor transcript using 1c (I.3) were detected in human fetal liver, but in adult liver the major transcript using exon 1b (I.4) had completely disappeared, and only the minor transcript using exon 1c (I.3) was observed (46). In addition, results of various clinical and experimental studies suggest that the switching of these exons 1 may result in aromatase overexpression and subsequent increase of in situ estrogen production under the new promoter in human breast, endometrial, and ovarian cancers, which will be described in detail in the following sections.


   III. Aromatase in Breast Cancer
Top
 Abstract
 I. Introduction
 II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
 VI. Summary
 References
 
A. Introduction
Previous in vitro biochemical studies showed that more than 72% (2), 70% (55), and 63% (56) of resected human breast cancer specimens had aromatase activity comparable with or greater than that found in other tissues. Miller et al. (55) also detected estrogen biosynthesis in all (247/247) breast adipose tissue specimens obtained from patients with breast cancer. In addition, O’Neill et al. (57) and Bulun et al. (58) both demonstrated that aromatase activity and expression in breast adipose tissues are highest in regions proximal to tumor. Bulun and co-workers (58, 59) also reported that the highest transcript levels of aromatase mRNA were found in the quadrant where the tumor was located, using competitive RT-PCR. We also demonstrated that the aromatase mRNA levels in the breast cancers were significantly increased compared with those in nonmalignant tissues (60). Together, these results indicate that increased aromatase activity and/or expression is associated with malignant phenotype or cancer in the human breast. However, it is well known that specimens of breast cancer tissue are composed of different cell types, such as adipocytes, stromal cells, infiltrating inflammatory cells, and carcinoma cells. In addition, diverse histological types of human breast carcinoma exist. Therefore, determination of aromatase activity and/or expression per unit weight of breast cancer tissue can result in underestimation of the levels of aromatase activity and/or expression in the cell types in which aromatase is expressed (61). These underestimations or false-negative results can also result in the misinterpretation that intratumoral or adipose tissue aromatase activity or expression in breast cancer is too low to sustain meaningful levels of intratumoral estrogen concentration (23). Therefore, to obtain a better understanding of intratumoral aromatization in human breast cancers, it is extremely important to correlate the morphological features of breast cancer with aromatization, i.e., to determine which cells are responsible for converting androgens to estrogens.

B. Localization
Biochemical studies including assays of tumor aromatase activity and tissue estrogen concentration have provided important and valuable information on the status of intratumoral aromatase or in situ estrogen production in human breast disorders, but it is nearly impossible to determine localization of aromatase in clinical specimens using these biochemical methods. The recent development of antibodies against aromatase has made it possible to examine the localization of aromatase in various tissues, including human ovary (12, 13, 62), placenta (63), and testis (64), as well as in their tumors (65, 66). Therefore, immunohistochemistry of aromatase in human breast disorders using antibodies against aromatase can provide important information on in situ estrogen metabolism in human breast tissues through localizing the possible sites of aromatization. Several groups of investigators have reported immunolocalization of aromatase in human breast disorders (67, 68, 69, 70, 71, 72, 73, 74). Results of these studies demonstrated overexpression of aromatase in breast cancer tissues, but somewhat different results have been reported on the localization of intratumoral aromatase. Sasano and associates (67, 68, 69, 70) have shown aromatase immunoreactivity both in adipocytes and stromal cells in breast carcinoma tissues (Fig. 3AGo). Intense immunoreactivity was detected in adipocytes located near carcinoma infiltration in almost all cases. Santner et al. (71) and Santen et al. (73) also demonstrated aromatase immunoreactivity predominantly in the stromal cells of breast carcinoma tissue. However, Esteban et al. (74) and Lu et al. (72) both reported aromatase immunoreactivity in epithelial or carcinoma cells and stromal cells in breast carcinoma tissues. We also obtained immunohistochemical localization of aromatase predominantly in the cytoplasm of carcinoma cells using the same monoclonal antibody that Lu et al. (72) employed (Fig. 3BGo), but immunoreactivity was weak and nuclear immunolocalization, possibly as a result of nonspecific reaction, was also detected. If the sites of aromatization are the epithelial or carcinoma cells, estrone, produced as a result of intratumoral aromatase, acts on carcinoma cells in an autocrine fashion. If these sites are in the stromal cells, estrone may act on carcinoma cells in a paracrine manner. Therefore, it is important to determine the exact sites of aromatization in human breast cancer to characterize intratumoral aromatase.



View larger version (125K):
[in this window]
[in a new window]
  		Figure 3. A, Immunohistochemical localization of aromatase in human breast carcinoma tissue using polyclonal antibody (invasive ductal carcinoma, 56-yr- old patient). Immunohistochemistry was performed using a streptavidin-biotin-amplified method on 10% formalin-fixed and paraffin-embedded tissue specimens. Immunoreactivity was detected in stromal cells in carcinoma and adipose tissue adjacent to carcinoma (arrows). Carcinoma cells were immunohistochemically negative for aromatase (x150). B, Immunohistochemical localization of aromatase in human breast carcinoma tissue using monoclonal antibody (invasive ductal carcinoma, 60-yr-old patient). Immunostain was performed as described above. Immunoreactivity was detected in cytoplasm of carcinoma cells (arrows) (x150).

 
Bulun and colleagues (61) reported that the distribution of stromal cells significantly correlated with the distribution of aromatase gene transcript levels using competitive RT-PCR and morphometric analysis of the histological specimens using a computer-assisted image-processing program (61). It was further shown that aromatase expression in human breast is determined by the ratio of stromal cells or fibroblasts in the resected specimens (75). Biochemically measured enzyme activity significantly correlated only with the level of immunoreactivity detected in the stromal cells, although aromatase immunoreactivity was detected in both carcinoma and stromal cells of breast cancer specimens (73). Recently, Santner et al. (71), after isolating stromal cells in breast cancer specimens, demonstrated directly that aromatase is present predominantly in the stromal cells. These results all suggest that the sites of intratumoral aromatization in human breast carcinoma tissues are stromal cells in cancer tissues and adipocytes and stromal cells in adjacent adipose tissues in human breast cancer tissues (summarized in Fig. 4Go). However, the results of immunohistochemistry and mRNA in situ hybridization are easily influenced by various factors involved during the process of specimen preparations, i.e., the types and duration of fixatives employed, especially in the cases of clinical materials and procedures such as the types of antibodies or probes employed (76). Future investigation, including the introduction of reliable monoclonal antibodies against aromatase, will attempt to clarify the exact sites of intratumoral aromatization in clinical samples of breast cancer.



View larger version (22K):
[in this window]
[in a new window]
  		Figure 4. Putative mechanism of in situ estrogen production in human breast cancer. Estrone (E1) is produced by aromatase from androgens in circulation and converted to 17ß-estradiol (E2) in carcinoma cells by 17ß-HSD type 1 (17ß-HSD 1). E2 produced in situ exerts its effects on carcinoma cells through binding to ER.

 
C. Genetic regulation
Studies of genetic regulation of aromatase in breast cancer were first performed in adipose tissues of resected breast specimens (59, 60, 61). Breast adipose tissue has been demonstrated as a major source of estrogens in menopausal women with breast cancer, and aromatase activity or expression in breast adipose tissue was also significantly associated with the presence of tumors (23, 24, 59, 60, 61). Transcripts in human adipose tissue in general contain three different major 5'-termini derived from specific untranslated exons 1, corresponding to expression derived from the proximal promoter 1d (PII), its splice variant 1c (I.3), and distal promoter 1b (I.4) (58, 61, 76, 77). Harada et al. (46) reported that aromatase transcripts from breast adipose tissues from healthy controls demonstrated tissue-specific use of exon 1b (I.4) or fibroblast type, but those from three of five breast cancer patients showed a switch from exon 1b (I.4) to exons 1c (I.3) and 1d (PII), both gonadal types of alternative exons 1 of aromatase gene. Agarwal and colleagues (77) also reported similar findings, i.e., exon 1b (I.4)-specific transcripts were predominant in breast adipose tissue obtained from cancer-free women while exon 1d (PII) and exon 1c (I.3) were chiefly used in breast adipose tissue of cancer patients. It was further shown that exon 1b (I.4)-specific transcripts of aromatase mRNA were predominantly present in adipose tissue samples obtained from normal women or women without breast cancer regardless of the tissue site or the age of individuals (78). Zhou et al. (79) also reported similar findings. In breast cancer tissue itself, we have demonstrated that the levels of aromatase mRNA predominantly transcribed from exon 1b (I.4) were higher than in nonmalignant tissues. Despite the same utilization patterns of exon 1b (I.4) (60, 80), switching from 1b (I.4) to 1c (I.3) occurred in breast cancer tissues themselves. In addition, the cases of breast cancer in which 1c (I.3) was used tended to demonstrate higher aromatase mRNA expression than those in which 1c was not used (60, 80). These results all suggest that aromatase overexpression occurs as a result of alternative splicing of exons 1 or by using alternative transcriptional sites, i.e., stromal cells begin to utilize gonadal type exons 1c (I.3) and 1d (PII) instead of fibroblast type exon 1b (I.4), possibly as a result of malignant transformation of breast. It then becomes important to study possible factors involved in genetic regulation or this switching of alternative exons 1 of aromatase gene in human breast cancer.

Zhao et al. (81) recently reported that prostaglandin E2 (PGE2), which is produced and secreted by breast tumor epithelial cells, fibroblasts, and infiltrating macrophages, is the most potent factor that stimulates aromatase expression via cAMP and results in utilization of exon 1d (PII). In addition, various cytokines, such as insulin-growth factor types I and II, interleukin (IL)-6, and IL-1, have been shown to stimulate aromatase activity in breast tumor-derived fibroblasts in the presence of dexamethasone (82). Therefore, PGE2, the above-listed cytokines, and other factors associated with malignant transformation and/or carcinoma-stromal interaction may be involved in these alternative splicing patterns of aromatase exons 1 in human breast cancer, which subsequently results in overexpression of intratumoral aromatase. However, there have been no studies reported on the correlation between results of alternative splicing of aromatase gene exons 1 described above and aromatase activity itself. Sourdaine et al. (83) recently reported the discrepancy between aromatase activity and mRNA expression in human breast cancer tissues. Therefore, the correlation between alternative splicing of aromatase gene exons 1 and aromatase activity in human breast cancer tissues needs to be clarified by further studies.

D. Clinical and pathological correlation
Overexpression of intratumoral aromatase itself can provide unopposed estrogen stimulation, which may result in increased proliferation of carcinoma cells and subsequent aggressive biological behavior of breast cancer. Therefore, it then becomes important to study the possible correlation between intratumoral aromatase and clinicopathological parameters of the patients, e.g., cell proliferation, nodal status, and clinical outcome. However, we could not find any significant correlations between intratumoral aromatase expression (assessed by immunohistochemistry and the presence or absence of lymph node metastasis), carcinoma cell proliferation (examined by immunostain of Ki67, a marker of proliferative cell), and clinical stage of the patients at the time of mastectomy, histological carcinoma subtypes, and the menopausal status of the patients (67, 69). Lipton et al. (39) also showed that there was no relationship between aromatase activity and disease-free interval or survival in 127 patients with breast cancer. Silva et al. (56) reported a significant correlation only between aromatase activity and the histological grade of the breast cancer specimens but no correlation between aromatase activity and clinical outcome of the patients. These results all indicate that overexpression of intratumoral aromatase itself does not necessarily confer a growth advantage on human breast carcinoma cells. Therefore, intratumoral aromatase status should be correlated with other factors involved in intratumoral estrogen metabolism.

If estrogens produced as a result of intratumoral aromatization have any innate biological significance in development and/or biological behavior of breast cancer, the status of intratumoral aromatase should then be correlated with estrogen receptor (ER) positivity of carcinoma cells. However, results of such a correlation between aromatase activity or expression and ER status have been inconsistent (55, 67, 69, 74, 84). In our study of aromatase immunohistochemistry (67), aromatase immunoreactivity in the stromal and/or adipocytes of the breast cancer tissue was not necessarily distributed adjacent to ER-positive carcinoma cells, even in the cases in which ER and aromatase were detected in the same carcinoma specimens. The correlation between regulation of ER and aromatase expression in human breast cancer tissue has not been studied. Therefore, the mechanism of expression of ER in relation to in situ estrogen synthesis in breast cancer should be further studied because local production of estrogen is important only when the breast cancer tissue contains ER. In addition, the correlation between ER status and aromatase may be further clarified when the correlation between ER-ß (84, 85, 86), a newly characterized ER, and intratumoral aromatase is examined. An increasing number of aromatase inhibitors are being introduced into clinical practice (87, 88), and these inhibitors can act on breast carcinoma through suppressing intratumoral aromatase activity. Newly developed aromatase inhibitors are specific and associated with minimum side effects (87, 88). Therefore, aromatase inhibitors are expected to be incorporated into postoperative adjuvant endocrine treatment of patients with breast carcinoma in the near future. The analysis of intratumoral aromatase and ER{alpha} and/or ERß status in resected specimens of breast carcinoma may therefore contribute to prediction of clinical response to aromatase inhibitors as ER status does in tamoxifen treatment.

17ß-HSD type 1 is another important factor of intratumoral estrogen metabolism that provides a growth advantage for tumor cells. In contrast to aromatase, both Poutanen et al. (10) and our group (69) demonstrated the expression of 17ß-HSD type 1 in carcinoma cells (Fig. 5Go). Aromatase and 17ß-HSD type 1 are not necessarily expressed in the same carcinoma specimens (69). However, among invasive carcinoma, invasive lobular carcinoma coexpressed these two enzymes in carcinoma tissues more frequently than invasive ductal carcinoma (69). There have been no studies on the correlation between intratumoral aromatase and estrogen sulfatase in human breast carcinoma tissues.



View larger version (99K):
[in this window]
[in a new window]
  		Figure 5. Immunohistochemistry of 17ß-HSD type 1 in invasive ductal carcinoma (51-yr-old patient). 17ß-HSD type 1 immunoreactivity was detected in carcinoma cells (arrows) (x300). [Reproduced with permission from H. Sasano et al.: J Clin Endocrinol Metab 81:4042–4046, 1996 (69 ). © The Endocrine Society.]

 
E. Male breast cancer
Male breast cancer is a rare malignant disease, accounting for 0.1% of all male cancer deaths in the United States each year (89, 90); its incidence is relatively lower in Japanese men as is the frequency of breast cancer in Japanese women (91). There has been strong circumstantial evidence to implicate hormonal factors, especially abnormal estrogen metabolism, in the development of male breast cancer (92, 93). ER has been detected more frequently in male breast cancer than in the female counterpart (93, 94). However, there has been no consistent evidence of increased serum estrogen concentrations or other estrogen abnormalities in men with breast cancer (89, 95). Therefore, despite the rarity of the disease, there has been considerable interest in the endocrine profiles of patients with male breast cancer.

We recently studied aromatase expression in male breast carcinoma (15 cases) and gynecomastia (30 cases) (68). Intratumoral aromatase overexpression was detected in all cases of carcinoma (Fig. 6Go) but in only 11/30 (37%) of gynecomastia. These results also indicated that increased intratumoral aromatase expression is considered to contribute to the increment of the in situ estrogen concentration. The high incidence of ER positivity described above and intratumoral aromatase overexpression in male breast carcinoma are therefore considered to provide a growth advantage for this estrogen-dependent tumor in a male environment characterized by relatively low serum levels of estrogen, which are hostile to tumor growth of breast carcinoma cells.



View larger version (119K):
[in this window]
[in a new window]
  		Figure 6. Immunohistochemical localization of aromatase in male breast carcinoma (69-yr-old patient). Aromatase immunoreactivity designated by arrows was detected in the stromal cells (S) but not in carcinoma cells (C) (magnification x150). [Reproduced with permission from H. Sasano et al.: J Clin Endocrinol Metab 81:3063–3067, 1996 (68 ). © The Endocrine Society.]

 

   IV. Aromatase in Endometrial Cancer
Top
 Abstract
 I. Introduction
 II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
V. Aromatase in Ovarian...
 VI. Summary
 References
 
A. Introduction
Endometrial carcinoma has become the most common invasive neoplasm of the female genital tract in the United States (96). In addition, its incidence has increased in Japan, and migration studies of Japanese women showed an increase in the rates of endometrial carcinoma in first- and second-generation women born in California (97). Based on several lines of epidemiological and clinicopathological evidence (96, 98, 99, 100, 101), there are two different forms of endometrial cancer, a low-grade neoplasm that is estrogen-related and occurs in younger, perimenopausal women, and a second, more virulent form, unrelated to estrogenic stimulation, that occurs in older postmenopausal women. The former type is termed "endometrioid endometrial carcinoma" in which estrogen is considered to play an important role in its development and subsequent biological behavior, and endometrial hyperplasia, especially atypical hyperplasia, can be its precursor (101). The latter type of endometrial cancer is designated "nonendometrioid endometrial carcinoma" and is generally associated with an atrophic endometrium and not with evidence of estrogenic stimulation (102). Nonendometrioid endometrial carcinoma includes high-grade tumors such as serous papillary carcinoma and clear cell carcinoma (96). As is expected, ER is generally positive in the great majority of endometrioid endometrial carcinomas (96, 98, 101) but not so in nonendometrioid endometrial carcinoma except for differentiated areas of the tumor (103). Therefore, the possible correlations between development of endometrioid endometrial carcinoma and abnormal estrogen metabolism have been examined by numerous investigations. However, although estrogen has been involved as a promoting factor in endometrioid endometrial cancer, there has been no consistent evidence of increased serum estrogen concentrations or other apparent systemic abnormalities of estrogens in women with endometrioid endometrial cancer (18, 104, 105), as observed in patients with breast cancer.

Aromatase activity has been reported in human endometrial cancer (32, 33, 106). Aromatase mRNA expression was also reported by Bulun et al. (107) and our group (108). Aromatase activity or expression has not been detected in normal endometrium (107, 109, 110, 111), although aromatase mRNA expression was reported in endometriosis and adenomyosis (109, 110, 111, 112, 113). Endometrial carcinoma tissue is composed of different types of cells such as epithelial or carcinoma cells and stromal cells. It is therefore important to localize the sites of aromatization in endometrial cancer, as in breast cancer.

B. Localization
Watanabe et al. (114) reported marked aromatase immunoreactivity in stromal cells of 28/42 cases (66.7%) of endometrioid endometrial carcinoma but none in normal or hyperplastic endometrium including atypical hyperplasia (114). In situ hybridization studies also revealed that mRNA hybridization signals of aromatase accumulate in the stromal cells but not in carcinoma cells (Fig. 7Go) (114). In addition, the distribution of aromatase mRNA correlated well with the immunohistochemical localization of aromatase. Marked aromatase expression was detected at the sites of frank invasion including myometrial invasion both at protein and mRNA levels (114). The presence of stromal invasion is the only reliable criteria for differentiating endometrial hyperplasia from endometrioid endometrial carcinoma (96, 98). No immunoreactivity or mRNA hybridization signals were detected in endometrial hyperplasia, including atypical hyperplasia, a putative precursor of endometrioid endometrial carcinoma associated with marked nuclear atypia of hyperplastic epithelial cells, but not with stromal invasion. Therefore, intratumoral aromatase in human endometrial proliferative lesions is considered to be associated with stromal invasion and to be expressed during the process of carcinoma-stromal interaction.



View larger version (125K):
[in this window]
[in a new window]
  		Figure 7. In situ hybridization of aromatase mRNA in endometrioid endometrial carcinoma. In situ hybridization was performed using the 27-base aromatase oligonucleotide probe corresponding to 847–873 (5'-GCGCATGACCAAGTCCAC GACAGGCTG-3') radiolabeled with 35S. A, accumulation of aromatase mRNA hybridization signals appearing as black dots on autoradiogram was detected in the stromal cells (S), but not in carcinoma cells (C). B, Negative control with a sense oligonucleotide probe with no detectable specific mRNA hybridization (x200). [Reproduced with permission from K. Watanabe et al.: Am J Pathol 146:491–500, 1995 (114 ).]

 
C. Genetic regulation
Genetic regulation of aromatase has not been extensively studied in endometrial carcinoma, in contrast to that in breast carcinoma described above. Bulun et al. (107) detected varying levels of aromatase transcripts in all eight cases of endometrial cancer examined. Bulun et al. also examined the untranslated first exons in 5'-termini of aromatase transcripts and reported that exon 1d (PII) and 1c (I.3.), both gonadal types of alternative exons 1 of aromatase gene, were primarily used (107). However, in our study, exon 1d (PII) was primarily used in three cases in which aromatase overexpression was not detected, but in the two cases in which the fibroblast type exon 1b (I.4) was used with other exons 1 as minor transcript, aromatase overexpression was demonstrated, employing both RT-PCR and immunohistochemical studies (108). Therefore, the possible involvement of alternative splicing, as well as use of multiple exon 1 transcripts in the overexpression of aromatase in human endometrial cancer, remains an unresolved issue because the number of endometrial carcinoma specimens examined in our study (108), as well as in the study of Bulun et al. (107), was limited.

Potential factors regulating the expression of intratumoral aromatase have not been characterized in human endometrioid endometrial cancer. However, considering the strong association of intratumoral aromatase expression with stromal invasion in endometrioid endometrial carcinoma described above, prostaglandins (81) or various cytokines (85, 115) that are derived from tumor-infiltrating macrophages or other inflammatory cells may also be involved in regulation of intratumoral aromatase expression in endometrial cancer tissues. Noble et al. (112) recently examined the effects of IL-1ß, IL-2, IL-6, IL-11, IL-15, tumor necrosis factor {alpha}, and PGE2 on aromatase expression in endometriosis-derived stromal cells. They demonstrated that only PGE2 stimulated aromatase activity in these cells as in breast cancer tissues (79). In addition, the majority of aromatase transcripts in PGE2-stimulated endometriosis-derived stromal cells contained specific sequences of gonadal type exon 1d (PII), whereas very few transcripts contained exons 1b (I.4)- or 1c (I.3)-specific sequences (112). Therefore, PGE2, possibly derived from infiltrating macrophages, lymphocytes, and carcinoma cells, may also stimulate expression of intratumoral aromatase of human endometrioid endometrial cancer as in the case of breast cancer.

D. Clinical and pathological correlation
Watanabe et al. (114) reported no correlation in endometrial carcinoma between aromatase expression or activity and clinicopathological factors such as clinical stage or histological grade and between aromatase expression or activity and steroid receptor status. Both intratumoral aromatase expression and activity tended to be higher in postmenopausal patients than in premenopausal patients, but differences did not reach statistical significance. Bulun et al. (107) reported no significant correlations between aromatase mRNA transcript levels and histological grade of the tumor, myometrial invasion, stage of the disease, or patient age. Therefore, intratumoral aromatase overexpression is much more frequently detected in endometrioid endometrial carcinoma than in breast carcinoma, but its significance also needs to be clarified by further studies including the possibility of application of aromatase inhibitors as one form of endocrine treatment of endometrial cancer (116).

17ß-HSD type I expression was also reported in human endometrium during the menstrual cycle (117) and in endometrial cancer (118). In contrast to aromatase, 17ß-HSD type I expression was also detected in normal endometrium. The enzyme appeared in surface epithelium and glandular cells during the early and midluteal phase of the menstrual cycle. In endometrial carcinoma, 17ß-HSD type I immunoreactivity was detected in 48% of the specimens (118). These results indicated that 17ß-estradiol can also be produced in human endometrioid endometrial cancer tissues, as reported in breast cancer. However, the correlation between aromatase and 17ß-HSD type I has not been examined in the same specimens of endometrial cancer. Yamamoto et al. (106) reported that aromatase and estrone sulfatase activities in endometrial carcinoma tissues were significantly higher than in normal endometrium, but estrone sulfotransferase activity was not different between normal endometrium and endometrial carcinoma (106). Therefore, estrone sulfatase may also be involved in providing biologically active estrogen in situ in human endometrial carcinoma.


   V. Aromatase in Ovarian Cancer
Top
 Abstract
 I. Introduction
 II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
VI. Summary
 References
 
A. Introduction
Placenta and ovary are the two major organs in which aromatase is abundantly expressed, and regulation of its gene expression and localization has been extensively studied (42, 47, 119). In normal cycling human ovary, aromatase is expressed in only one follicle, the dominant follicle, during folliculogenesis both at the protein and mRNA levels (13, 62, 120); ER is also expressed in the dominant follicle during folliculogenesis (121). However, several groups reported aromatase expression sporadically in follicular theca cells (122, 123, 124). Aromatase expression was also detected in luteinized granulosa cells of one corpus luteum per patient in the midproliferative to the premenstrual phase and disappeared in the menstrual to the early proliferative phase (13, 62, 120). 17ß-HSD type 1 is also expressed in the dominant follicle (125). Therefore, 17ß-estradiol locally produced in a selected follicle is considered to have important roles in the process of follicular maturation and growth as a local regulation in human ovary. It is therefore not surprising that some cases of granulosa cell tumor, which is the second most common sex cord-stromal tumor of the ovary after fibroma and is well known to be associated with estrogen secretion or clinical hyperestrogenic manifestations (120, 126, 127, 128), express aromatase, especially in tumor cells with clear cytoplasm (129). These cells in a granulosa cell tumor were also positive for Ad4-binding protein (Ad4BP) or steroidogenic factor-1, a transcription factor that regulates the expression of the steroidogenic cytochrome P450 genes (130). In addition, Bulun et al. (131) reported that exon 1d or PII was used in the expression of aromatase gene in a case of ovarian granulosa cell tumor. However, these sex-cord-stromal tumors of the ovary, including granulosa cell tumors, which comprise approximately 8% of all ovarian tumors, can be estrogen-producing, but by no means estrogen-dependent, neoplasms (132).

Common epithelial tumors or surface epithelial-stromal tumors of the ovaries account for 60% of all ovarian neoplasms and 80–90% of primary ovarian malignancies (133, 134). Results from several case-control and cohort studies strongly suggest that endocrine factors play an important role in these ovarian cancers (135). High levels of gonadotropins in women in early postmenopause have been postulated to play a role in the development of epithelial ovarian neoplasms (135, 136). However, there has been no conclusive evidence regarding a correlation between serum estrogen levels or other systemic sex steroid abnormalities and the development of common epithelial ovarian malignancies (15, 16, 135, 137). In particular, MacDonald et al. (20) have reported that elevated levels of androstenedione, the precursor of estrogen, may have an important role in the development of ovarian cancer in the postmenopausal years. Aromatase activity and steroid receptors have been demonstrated in these epithelial ovarian cancers as described previously (34, 35, 36). Thus, intratumoral aromatase may be important in the development and/or biological behavior of surface epithelial-stromal ovarian neoplasms.

B. Localization
In our recent study of ovarian surface epithelial-stromal tumors, aromatase immunoreactivity was observed in stromal cells in 35 of 44 (79.5%) ovarian carcinomas, 3 of 7 carcinomas of low malignant potential, and none of 14 adenomas (138). The high incidence of aromatase expression in carcinomas and its absence in benign adenomas is consistent with patterns of aromatase expression in other estrogen-dependent human neoplasms, including breast and endometrial malignancies described previously. In surface epithelial-stromal tumor of the ovary, carcinoma of low malignant potential or atypical proliferating tumors that are associated with increased proliferation of tumor cells, but not with stromal invasion, demonstrated aromatase immunoreactivity in the stromal cells adjacent to the carcinoma (138), in contrast to endometrial lesions in which aromatase expression was detected in well differentiated endometrioid carcinomas with stromal invasion but not in atypical endometrial hyperplasia without stromal invasion. However, pronounced aromatase immunoreactivity was also detected at the sites of frank invasion in ovarian carcinoma (138). The absence of aromatase expression in epithelial or carcinoma cells is also consistent with the absence of aromatase mRNA in isolated epithelial cells of ovarian carcinoma (139) and the presence of Ad4BP or steroidogenic factor-1 in stromal cells but not in carcinoma cells in ovarian epithelial malignancy (140). However, Kitawaki et al. (141) immunolocalized aromatase in the cytoplasm of neoplastic cells in both benign and malignant ovarian epithelial tumors.

C. Genetic regulation
We examined alternative utilization of multiple copies of exons 1 in 11 ovarian epithelial-stromal carcinomas (138). The transcripts using exons 1c (I.3) and 1d (P II), both gonadal types of alternative exons 1 of aromatase gene, were detected in four and five cases of carcinoma, respectively. Expression of aromatase in nonneoplastic human ovary has been demonstrated to use mainly exon 1c (I.3) or 1d (PII). Our study demonstrated that exons 1c (I.3) and 1d (II) were detectable in 10 of 12 patients in which aromatase mRNA was detected. One case used three different varieties of exon 1 as the major transcripts, and two cases used two different types of exon 1. However, patterns of exon 1 utilization were not necessarily correlated with in situ aromatase overexpression in human epithelial stromal ovarian carcinomas (138). In addition, there was no correlation between the level of aromatase activity and the aromatase-labeling index or aromatase mRNA in ovarian carcinoma cases examined, as Sourdaine et al. (83) reported in breast cancer. There was, however, a good correlation between the labeling index or areas of stromal cells positive for aromatase determined by computer-assisted image analysis and the amount of aromatase mRNA in ovarian carcinomas. Aromatase activity is regulated by the number of aromatase molecules rather than by changes in the catalytic ability of each molecule (142, 143). However, aromatase activity, assessed by the tritiated water method, is also determined by the amount of both aromatase enzyme and NADPH-cytochrome P450 reductase as described previously (144). Therefore, the discrepancy between aromatase expression at both mRNA and protein levels and aromatase activity in our study (138) and that of Sourdaine et al. (83) may be due to differences in the relative ratio of the aromatase cytochrome P450 and NADPH-cytochrome P450 reductase among the cases examined.

D. Clinical and pathological correlation
Intratumoral aromatase expression and activity were not correlated with ages of the patients, but aromatase immunointensity and aromatase-positive regions were significantly higher in serous adenocarcinomas than mucinous adenocarcinomas in our study (138). Slotman et al. (35) reported correlation of high tumor progesterone receptor levels with longer survival of the patients but no significant correlation between aromatase activity and the prognosis of the patients. Noguchi et al. (36, 145) and Kitawaki et al. (143) both demonstrated aromatase activity in surface epithelial-stromal ovarian carcinoma and indicated that aromatase in these tumors of the ovary is significantly correlated with progesterone receptor levels, although the biological significance of this observation remains unknown. 17ß-HSD type I expression was detected in 4 of 8 cases of carcinoma of low malignant potential and 20 of 30 cases of invasive carcinoma but none of benign cystadenoma (125). Patterns of expression of 17ß-HSD type 2 among epithelial-stromal ovarian tumors, including correlation with malignant phenotype, also suggest the possible roles of 17ß-HSD-type 1 in in situ estrogen production in ovarian carcinomas.

In contrast to human breast and endometrial carcinomas, evaluation of intratumoral aromatase in ovarian carcinoma is complicated by the presence of stromal luteinization or enzymatically active stromal cells (EASCs) (146). EASCs are detected frequently in ovarian stroma adjacent to space occupying lesions of the ovary including epithelial-stromal tumors, both benign and malignant and metastatic tumors (147). Transformation of stromal cells into EASCs is considered as one characteristic of ovarian stroma and has been postulated to be due to substances produced by the tumor cells (148) or to the pressure of the tumor on adjacent tissue, because EASCs are detected not only in primary ovarian tumor but also in metastatic tumor (146). However, the great majority of EASCs expressed Ad4BP, P450scc (cholesterol side-chain cleavaging enzyme), 3ß-HSD, and P450c17 (17{alpha}-hydroxylase) but not aromatase, as in luteinized cells detected in thecoma, hyperthecosis, and polycystic ovaries (62, 65, 128, 129, 150, 151). Therefore, androgens rather than estrogens are produced in these functioning stroma or EASCs (62). Hyperestrogenic manifestations have been reported in a variety of common epithelial tumors of the ovary (20, 137, 152). These hyperestrogenic manifestations have been considered to be associated with both intratumoral (153) and extratumoral (154) luteinized stromal cells. These luteinized stromal cells have been considered to contribute to hyperestrogenic manifestations by producing androgens and subsequent peripheral or intraovarian conversion to estrogens (12). Among different histological types of common epithelial ovarian tumors, EASCs have been more frequently detected in mucinous ovarian tumors (146, 147). On the other hand, aromatase expression is more frequently detected in stromal cells of serous carcinomas than in mucinous carcinomas (138). Aromatase expression in stromal cells of the human ovary, which may result from various factors produced by carcinoma cells, is therefore considered to arise differently from the development of intra- or extratumoral EASCs in terms of its mechanism and hormonal metabolism.

These results indicate that not only intratumoral aromatase, but also intratumoral and extratumoral functioning stromal cells, should be considered when evaluating biological and clinical significance of in situ estrogen metabolism in ovarian carcinoma.


   VI. Summary
Top
 Abstract
 I. Introduction
 II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
 VI. Summary
References
 
In human estrogen-dependent neoplasms such as breast, endometrioid endometrial, and surface epithelial-stromal ovarian carcinomas, intratumoral aromatase is considered to play important roles in converting circulating androgens derived from adrenal cortex and/or ovary to estrogens, possibly in association with 17ß-HSD type 1 and estrogen sulfatase. Analysis of intratumoral aromatase in these estrogen-dependent neoplasms is important not only in understanding the development and biological behavior of these tumors, but also in the clinical management of these patients, because suppression of intratumoral aromatase by newly developed aromatase inhibitors may provide new potentials in endocrine therapy of these patients.


   Footnotes
 
Address reprint requests to: Hironobu Sasano, M.D., Department of Pathology, Tohoku University School of Medicine, 2–1 Seiryou-machi, Aobu-ku, Sendai 980-8575, Japan.

1 This work was supported in part by Public Trust Haraguchi Memorial Cancer Research Fund, Tokyo, Japan, The Grant-in-Aid for Cancer Research 7–1 from the Ministry of Health and Welfare, Japan, and a grant from the Ministry of Education, Japan. Back


   References
Top
 Abstract
 I. Introduction
 II. Aromatase Cytochrome P450...
 III. Aromatase in Breast...
 IV. Aromatase in Endometrial...
 V. Aromatase in Ovarian...
 VI. Summary
 References
 

  1. Henderson IC, Canellos GP 1990 Cancer of the breast: the past decade. N Engl J Med 302:17–30[Medline]
  2. Miller WR 1991 Aromatase activity in breast tissue. J Steroid Biochem Mol Biol 39:783–790[CrossRef]
  3. Schweikert HU, Milewich L, Wilson JD 1976 Aromatization of androstenedione by cultured human fibroblasts. J Clin Endocrinol Metab 43:785–795[Abstract]
  4. Longcope C, Pratt JH, Schneider SH, Fineberg SE 1978 Aromatization of androgens by muscle and adipose tissue in vivo. J Clin Endocrinol Metab 46:146–152[Abstract]
  5. Sasano H, Uzuki M, Sawai T, Nagura H, Matsunaga G, Kashimoto O, Harada N 1997 Aromatase in human bone tissue. J Bone Miner Res 12:1416–1423[CrossRef][Medline]
  6. Hobkirk R 1993 Steroid sulfation. Trends Endocrinol Metab 4:69–74
  7. Reed MJ, Purohit A 1993 Sulphatase inhibitors. The rationale for the development of a new endocrine therapy. Rev Endocr Relat Cancer 45:51–62
  8. Reed MJ, Purohit A, Howarth NM, Potter BVL 1994 Steroid sulphatase inhibitors: a new endocrine therapy. Drugs Future 19:6734–6738
  9. Peltoketo H, Isomaa V, Maentausta O, Vihko R 1988 Complete amino acid sequence of human placental 17ß-hydroxysteroid dehydrogenase deduced from cDNA. FEBS Lett 239:73–77[CrossRef][Medline]
  10. Poutanen M, Isomaa V, Lehto VP, Vihko R 1992 Immunological analysis of 17ß-hydroxysteroid dehydrogenase in benign and malignant human breast tissue. Int J Cancer 50:386–390[Medline]
  11. Poutanen M, Isomaa V, Peltoketo H, Vihko R 1995 Role of 17ß-hydroxysteroid dehydrogenase type 1 in endocrine and intracrine estradiol biosynthesis. J Steroid Biochem Mol Biol 55:525–532[CrossRef][Medline]
  12. Sasano H, Sasano N 1989 What’s new in the localization of sex steroids in the human ovary and its tumors? Pathol Res Pract 185:942–948[Medline]
  13. Sasano H, Okamoto M, Mason JI, Simpson ER, Mendelson CR, Sasano N, Silverberg SG 1989 Immunolocalization of aromatase, 17 alpha-hydroxylase and side-chain-cleavage cytochromes P-450 in the human ovary. J Reprod Fertil 85:163–169[Abstract/Free Full Text]
  14. Lipworth L, Adami HO, Trichopoulos D, Carlstrom K, Mantzoros C 1996 Serum steroid hormone levels, sex hormone-binding globulin, and body mass index in the etiology of postmenopausal breast cancer. Epidemiology 7:96–100[Medline]
  15. Dorgan JF, Stanczyk FZ, Longcope C, Stephenson Jr HE, Chang L, Miller R, Franz C, Falk RT, Kahle L 1997 Relationship of serum dehydroepiandrosterone (DHEA), DHEA sulfate, and 5-androstene-3 beta, 17 beta-diol to risk of breast cancer in postmenopausal women. Cancer Epidemiol Biomarkers Prev 6:177–181[Abstract]
  16. Berrino F, Muti P, Micheli A, Bolelli G, Krogh V, Sciajno R, Pisani P, Panico S, Secreto G 1996 Serum sex hormone levels after menopause and subsequent breast cancer. J Natl Cancer Inst 88:291–296[Abstract/Free Full Text]
  17. James VHT, Reed MJ, Folkerd EJ 1981 Studies of oestrogen metabolism in postmenopausal women with cancer. J Steroid Biochem 15:235–245[CrossRef][Medline]
  18. Helzlsouer KJ, Gordon GB, Alberg AJ, Bush TL, Comstock GW 1992 Relationship of prediagnostic serum levels of dehydroepiandrosterone and dehydroepiandrosterone sulfate to the risk of developing premenopausal breast cancer. Cancer Res 52:1–4[Abstract/Free Full Text]
  19. Cramer DW, Welch WR 1983 Determinants of ovarian cancer risk. II. Inferences regarding pathogenesis. J Natl Cancer Inst 71:717–721
  20. MacDonald PC, Grodin JM, Edman CD, Vellios F, Siiteri PK 1976 Origin of estrogen in a postmenopausal woman with a nonendocrine tumor of the ovary and endometrial hyperplasia. Obstet Gynecol 47:644–650[Abstract/Free Full Text]
  21. Jeppsson S, Karlsson S, Kullander S 1986 Gondal steroids, gonadotropins and endometrial histology in postmenopausal women with malignant ovarian tumors. Acta Obstet Gynecol Scand 65:207–210[Medline]
  22. Judd HL, Davidson BJ, Frumar AM, Shamonki IM, Lagasse LD, Ballon SC 1980 Serum androgens and estrogens in postmenopausal women with and without endometrial cancer. Am J Obstet Gynecol 136:859–871[Medline]
  23. Miller WR, Hawkins RA, Forrest AP 1982 Significance of aromatase activity in human breast cancer. Cancer Res 42[Suppl]:3365s–3368s
  24. Perel E, Wilkins D, Killinger DW 1980 The conversion of androstenedione to estrone, estradiol, and testosterone in breast tissue. J Steroid Biochem 13:89–94[CrossRef][Medline]
  25. Reed MJ, Owen AJ, Lai LC, Coldham NG, Ghilchik MW, Shaikh NA, James VHT 1989 In situ oestrogen synthesis in normal breast and breast tumour tissues: effect of treatment with 4-hydroxyandrostenedione. Int J Cancer 44:233–237[Medline]
  26. Bradlow HL 1982 A reassessment of the role of breast tumor aromatase. Cancer Res 42[Suppl]:3382s–3386s
  27. Thorsen T, Tangen M, Stoa KF 1982 Concentration of endogenous oestradiol as related to oestradiol receptor sites in breast tumor cytosol. Eur J Cancer Clin Oncol 18:333–337[CrossRef][Medline]
  28. van Landeghem AAJ, Portman J, Mabauurs M 1985 Endogenous concentration and subcellular distribution of estrogens in normal and malignant human breast tissue. Cancer Res 45:2900–2906[Abstract/Free Full Text]
  29. Yue W, Wang JP, Hamilton CJ, Demers LM, Santen RJ 1998 In situ aromatization enhances breast tumor estradiol levels and cellular proliferation. Cancer Res 58:927–932[Abstract/Free Full Text]
  30. de Jong PC, van de Ven J, Nortier HW, Maitimu-Smeele I, Donker TH, Thijssen JH, Slee PH, Blankenstein RA 1997 Inhibition of breast cancer tissue aromatase activity and estrogen concentrations by the third-generation aromatase inhibitor vorozole. Cancer Res 57:2109–2111[Abstract/Free Full Text]
  31. Bezwoda WR, Mansoor N, Dansey R 1987 Correlation of breast tumour aromatase activity and response to aromatase inhibition with aminoglutethimide. Oncology 44:345–349[Medline]
  32. Tseng L, Mazella J, Mann WJ, Chumas J 1982 Estrogen synthesis in normal and malignant human endometrium. J Clin Endocrinol Metab 55:1029–1031[Abstract]
  33. Yamaki J, Yamamoto T, Okada H 1985 Aromatization of androstenedione by normal and neoplasmic endometrium of the uterus. J Steroid Biochem 22:63–66[CrossRef][Medline]
  34. MacLusky NJ, Voit R, Lazo JS, Schwartz PE, Merino MJ, Eisenfeld AE, Naftolin F 1987 Aromatase activity in human ovarian cancer. Steroids 50:423–433[CrossRef][Medline]
  35. Slotman BJ, Kuhnel R, Rao BR, Dijkhuizen GH, de Graaff J, Stolk JG 1989 Importance of steroid receptor and aromatase activity in the prognosis of ovarian cancer: high tumor progesterone receptor levels correlate with longer survival. Gynecol Oncol 33:76–81[CrossRef][Medline]
  36. Noguchi T, Kitawaki J, Tamura T, Kim T, Kanno H, Yamamoto T, Okada H 1993 Relationship between aromatase activity and steroid receptor levels in ovarian tumors from postmenopausal women. J Steroid Biochem Mol Biol 44:657–660[CrossRef][Medline]
  37. Pasqualini JR, Chetrite G, Blacker C, Feinstein MC, Delalonde L, Talbi M, Maloche C 1996 Concentrations of estrone, estradiol, and estrone sulfate and evaluation of sulfatase and aromatase activities in pre- and postmenopausal breast cancer patients. J Clin Endocrinol Metab 81:1460–1464[Abstract]
  38. Bolufer P, Ricart E, Lluch A, Vazquez C, Rodriguez A, Ruiz A, Llopis F, Garcia-Conde J, Romero R 1992 Aromatase activity and estradiol in human breast cancer: its relationship to estradiol and epidermal growth factor receptor and to tumor-node-metastasis staging. J Clin Oncol 10:438–446[Abstract/Free Full Text]
  39. Lipton A, Santen RJ, Santner SJ, Harvey HA, Sanders SI, Matthews YL 1992 Prognostic value of breast cancer aromatase. Cancer 70:1951–1955[CrossRef][Medline]
  40. Labrie F 1991 Intracrinology. Mol Cell Endocrinol 78:C113–C118
  41. Labrie F, Belanger A, Simard J, Luu-The V, Labrie C 1995 DHEA and peripheral androgen and estrogen formation: intracrinology. Ann NY Acad Sci 774:16–28[Medline]
  42. Simpson ER, Mahendroo MS, Means GD, Kilgore MW, Hinshelwood MM, Graham-Lorence S, Amarneh B, Ito Y, Fisher CR, Michael MD, Mendelson CR, Bulun SE 1994 Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis. Endocr Rev 15:342–355[CrossRef][Medline]
  43. Bulun SE, Simpson ER, Word RA 1994 Expression of the CYP19 gene and its product aromatase cytochrome P450 in human uterine leiomyoma tissues and cells in culture. J Clin Endocrinol Metab 78:736–743[Abstract]
  44. Harada N, Yamada K, Saito K, Kibe N, Dohmae S, Takagi Y 1990 Structural characterization of the human estrogen synthetase (aromatase) gene. Biochem Biophys Res Commun 166:365–372[CrossRef][Medline]
  45. Means GD, Mahendroo MS, Corbin CJ, Mathis JM, Powell FE, Mendelson CR, Simpson ER 1989 Structural analysis of the gene encoding human aromatase cytochrome P-450, the enzyme responsible for estrogen biosynthesis. J Biol Chem 264:19385–19391[Abstract/Free Full Text]
  46. Harada N, Utsumi T, Takagi Y 1993 Tissue-specific expression of the human aromatase cytochrome P-450 gene by alternative use of multiple exons 1 and promoters, and switching of tissue-specific exons 1 in carcinogenesis. Proc Natl Acad Sci USA 90:11312–11316[Abstract/Free Full Text]
  47. Means GD, Kilgore MW, Mahendroo MS, Mendelson CR, Simpson ER 1991 Tissue-specific promoters regulate aromatase cytochrome P450 gene expression in human ovary and fetal tissues. Mol Endocrinol 5:2005–2013[CrossRef][Medline]
  48. Mahendroo MS, Means GD, Mendelson CR, Simpson ER 1991 Tissue-specific expression of human P-450AROM: the promoter responsible for expression in adipose is different from that utilized in placenta. J Biol Chem 266:11276–11281[Abstract/Free Full Text]
  49. Honda S, Harada N, Takagi Y 1994 Novel exon 1 of the aromatase gene specific for aromatase transcripts in human brain. Biochem Biophys Res Commun 198:1153–1160[CrossRef][Medline]
  50. Harada N 1992 A unique aromatase (P450AROM) mRNA formed by alternative use of tissue-specific exons 1 in human skin fibroblasts. Biochem Biophys Res Commun 189:1001–1007[CrossRef][Medline]
  51. Zhou C, Zhou D, Esteban J, Murai J, Siiteri PK, Wilczynski S, Chen S 1996 Aromatase gene expression and its exon 1 usage in human breast tumors. Detection of aromatase messenger RNA by reverse transcription-polymerase chain reaction. J Steroid Biochem Mol Biol 59