To understand the significance of estrogen receptor β (ERβ) in mammary carcinogenesis, we evaluated the expression of ERβ in preinvasive mammary tumors. The percentage of ERβ-positive epithelial or tumoral cells was assayed by quantitative immunohistochemistry using an image analyzer in 130 lesions of varying histological risk from 118 patients [71 with benign breast disease (BBD) and 59 with carcinoma in situ (CIS)] and compared with 118 adjacent histologically normal glands. Five groups of lesions with an increasing risk of invasive cancer, from BBD without hyperplasia to high-grade CIS, were studied. Results were compared with ERα and Ki67 immunostaining. The percentage of ERβ-positive cells was high (median, 85%) in “normal” mammary glands and in nonproliferative BBD and decreased significantly (P < 0.0001) in proliferative BBD without atypia and in CIS, contrasting with an inverse progression for the ERα level. In normal mammary glands, the ERβ level did not vary according to the nature of the lesion at the periphery and was significantly higher (P < 0.007) than in adjacent preinvasive lesions, except in nonproliferative BBD. The ERβ level decreased in proliferative BBD, anticipating the ERα increase, which was significant in BBD with atypia. In high-grade ductal carcinoma in situ, both ER levels were low. The ratio between ERβ and ERα was high in normal glands, and decreased significantly in proliferative lesions. ERβ staining was inversely correlated with Ki67 (r = −0.333; P < 0.001), more particularly in high-grade ductal carcinoma in situ (r = −0.57; P < 0.02). The marked and early decreased level of ERβ protein associated with other criteria of cell proliferation suggests a protective effect of ERβ against the mitogenic activity of estrogens in mammary premalignant lesions. Knowledge of the ERβ and ERα content in each preinvasive lesion should help to rationalize antiestrogen preventive therapy adapted to each individual patient.

The multistep model of carcinogenesis that has been well established for colorectal cancer (1) might also be valid in understanding the occurrence of breast cancer. However, the genes involved in mammary carcinogenesis are less clearly defined, with the exception of hereditary breast cancers. One approach for understanding which genes are altered first during mammary carcinogenesis is to analyze the expression of putative oncogenes and tumor suppressor genes in CIS3 and in precancerous lesions, such as high-risk BBD defined according to the histological criteria of Dupont and Page (2). The fact that high-risk BBD and CIS correspond to some early stages of invasive breast cancer is supported by the frequency of molecular alterations such as loss of heterozygosity at different loci in these lesions (3), suggesting alteration of tumor suppressor genes. Among the genes involved in sporadic human breast cancers, those related to estrogen action are good candidates because estrogens are known to be tumor promoters, possibly via their mitogenic activity (4). The mechanism of the promoter effect of estrogens in mammary cancer via their receptor has been reevaluated based on the recent discovery of a second ER (ERβ), first cloned in the rat prostate (5) but also expressed in various human tissues, including mammary glands and breast cancer from a gene located in 14q22-24 (6, 7). The presence of ERα, but not of ERβ, seems to be required to develop mammary duct and breast cancer, as shown by knock-out experiments in mice (8, 9). In vitro functional studies suggest that ERβ might modulate ERα action (10, 11). However, the first results, based mostly on the estimation of ERβ mRNA by reverse transcription-PCR analysis, have been controversial because ERβ has been suggested to be of both good (12, 13) and bad (14) prognostic significance in breast cancer.

The development of antibodies specific to human ERα (15) and more recently to ERβ (16) allows study of the expression of these two genes at the protein level in an attempt to understand their significance in mammary carcinogenesis and, possibly, to use them to help define adequate targeted preventive therapies. We previously showed in a population of preinvasive mammary lesions significantly increased ERα expression in BBD with atypia and in low-grade CIS with no correlation with cathepsin D expression (17).

In the same population of patients, we have now quantified the expression of ERβ in the nuclei of epithelial and cancer cells at five different stages as described by Dupont and Page (2) according to the risk of developing invasive breast cancer. The results were compared with ERα and Ki67 expression to determine whether the ERβ level varied in preinvasive lesions and how it varied compared with ERα.

Patient Characteristics.

Patients (n = 118) with BBD or in situ breast carcinoma (Table 1) were included in a multicenter prospective study from 1994 to 1998 at the University Hospitals of Montpellier and Nîmes, France. Written consent was obtained for all patients. Investigations were approved by the local ethics committee. In 61 patients, blood samples were collected for hormone assay on the day of surgery, and hormonal status was defined as described (17). Among the 55 menopausal patients, 11 received estrogen replacement therapy.

Breast Tissue Sampling.

All tissues were collected by breast surgery for diagnostic or therapeutic purposes from 60 patients with BBD and 58 patients with CIS (Table 1). BBDs were classified according to the relative risk of invasive breast cancer using the criteria of Dupont and Page (2) and DCIS according to the nuclear grade (18, 19). For each patient, formalin-fixed, paraffin-embedded tissue blocks containing the highest risk lesion and histologically normal glandular structures at the periphery of lesions were selected. When different histological lesions with the same risk were present, they were included, which explains why 130 lesions were analyzed in only 118 patients. Low- and intermediate-grade DCIS and LCIS samples were pooled for statistical analysis. One high-grade DCIS was associated with LCIS. A total of 3 atypical hyperplasias and 37 CISs were located at the periphery of invasive carcinomas. No significant difference was found for ERβ staining in these lesions compared with CIS and atypical hyperplasia without invasive carcinoma, allowing these lesions to be included in the study.

Immunohistochemistry.

ERβ immunohistochemical analysis was performed on each paraffin-embedded block (5-μm section) and on an adjacent section stained previously with the 1D5 ERα monoclonal antibodies (17, 20), using the chicken polyclonal ERβ503 IgY antibody (16). This antibody recognizes total ERβ proteins. Immunostaining was revealed by a streptavidin-biotin enhanced immunoperoxidase technique. After antigen retrieval by pressure cooking in 0.1 m EDTA buffer (pH 7.2) for 10 min (21) and endogenous peroxidase blocking with 1% H202, sections were incubated with normal donkey serum (1:40 dilution) in PBS for 30 min at 22°C. Sections were then incubated overnight at 4°C with ERβ 503 IgY (1:1250 dilution). The complex was revealed using antichicken biotinylated antibody (1:300 dilution; Vector, Compiègne, France) and streptavidin-peroxidase complex (1:1000 dilution; Dako, Trappes, France). 3,3′-Diaminobenzidine tetrahydrochloride (Sigma, Saint Quentin Fallavier, France) was used as chromogen, and the samples were counterstained with hematoxylin (Dako), dehydrated, and mounted. Tissue sections were washed with 0.1% Tween 20 in PBS between each immunostaining step. Immunostaining specificity was checked by preincubating ERβ 503 IgY with a 90-fold excess of recombinant human ERβ (Panvera, Madison, WI) or with nonspecific IgY (1:36000 dilution; Nordic, Tilburg, the Netherlands). In each experiment, sections of rat prostate and paraffin-embedded cell pellets (ERβ-positive cancer cell line) were used as positive external controls. Ki67 staining was performed using the MIB1 antibody (Immunotech, Marseilles, France) on 61 lesions (22). We used the same antigen retrieval procedure that was used for ERβ. Immunostaining was revealed using a streptavidin-biotin immunoperoxidase technique.

Immunostaining Quantification.

ERβ and Ki67 staining was quantified on structures corresponding to higher risk lesions and in adjacent histologically normal ducts and lobules with an image analyzer as described for ERα (17). Only nuclei were quantified following nuclear counterstaining and automatic selection by the analyzer. ERβ staining was expressed as a percentage of positive nuclei.

Statistical Analysis.

To compare the ERβ and ERα values between all groups, we used the Kruskal-Wallis test. When the test was significant, we used the Bonferroni correction to look for significance of subgroups. For paired samples, we used the Wilcoxon test for two samples. The correlation between the former variables was estimated using the Spearman test. The overall threshold of significance was 0.05.

ERβ Level in Benign Breast Lesions and CIS.

Fig. 1, a–c, shows the specificity of ERβ immunostaining using the 503 IgY antibody. This antibody gives strong immunoreactivity in the nuclei of epithelial cells and weak cytoplasmic staining. Nuclear staining was completely eliminated when the antibodies were preadsorbed with ERβ protein, whereas cytoplasmic staining persisted, suggesting a cross-reactivity with other cytoplasmic proteins. In this study, only nuclear staining in epithelial or tumoral cells was quantified. Interestingly, ERβ, but not ERα, was also present in the nuclei of cells other than epithelial cells, in both “normal” and tumoral structures, particularly in fibroblasts, lymphocyte infiltrates (6), macrophages, and in the endothelial cells of vessels (Fig. 1, e and f), as has also been described in rodents (16, 23). In the normal resting gland, ERβ was present both in luminal and myoepithelial cells (Fig. 1,e), whereas ERα was detected only in some luminal epithelial cells (Fig. 1,d). In most lesions, there was no parallelism between ERα and ERβ staining, as shown in Fig. 1, d, e, g, and h, which illustrate the case of a single patient. This dissociation also confirmed the absence of cross-reactivities between the two antibodies to ERα and to ERβ, as has been established in other studies (16, 24) and in cells transfected with only ERα or ERβ.4

The percentage of ERβ-positive cells (Fig. 2) was found to vary according to patient, but was much higher in normal mammary glands and in nonproliferative BBD and decreased significantly (P < 0.0001) in proliferative BBD without atypia and in CIS. Fig. 1 shows representative strong ERβ staining in a normal duct (Fig. 1,e) and cyst (Fig. 1,f) and its decrease in high-risk lesions such as proliferative BBD (Fig. 1,i) and DCIS (Fig. 1,h). Fig. 3 illustrates the ERβ level in each lesion with the corresponding normal adjacent glands. In normal mammary glands, the percentage of ERβ-positive cells was high and did not vary according to the nature of the lesion at the periphery, contrary to ERα-positive cells in the adjacent section (17). The ERβ level significantly decreased (P < 0.007) in all preinvasive lesions except in nonproliferative BBD compared with adjacent normal glands.

Comparison Between ERβ and ERα Levels.

Fig. 1 illustrates the dissociation between ERα (Fig. 1, d and g) and ERβ (Fig. 1, e and h) staining in a representative normal duct and in an intermediate-grade DCIS. The ERα level was studied previously on adjacent sections of ERβ (17), and the distribution of ERβ and of ERα nuclear staining was compared in normal mammary glands (at the periphery of the lesions) and in the five groups with increasing risk for invasive breast cancer (Fig. 4). In normal glands and in nonproliferative BBD, ERα levels were low (respective median values, 4 and 1.3%), whereas ERβ levels were high (respective median values, 85 and 79%). ERβ decreased in proliferative BBD, anticipating the ERα increase described previously (17) in proliferative BBD with atypia. In high-grade DCIS, both ERs were generally low or absent. Fig. 5 shows the ratio of the ERβ level (+ 1) to the ERα level (+ 1) in normal glands and in the preinvasive lesions. This ratio was high (>10) in normal glands and in nonproliferative BBD and then decreased as the risk of invasive cancer increased.

Other Correlations.

The Ki67 nuclear staining determined in adjacent sections increased with histological proliferation, whereas ERβ staining decreased (Table 2). Ki67 staining was negatively correlated with ERβ in both lesions and normal glands (r = −0.33; P < 0.001). The negative correlation was stronger when only the proliferative ducts with atypia and DCIS (r = −0.46; P < 0.01) or high-grade DCIS (r = −0.57; P < 0.02) were considered. No correlation was obtained between Ki67 and ERα staining (not shown).

In both lesions and normal glands, the percentage of ERβ-positive cells was also found to be negatively correlated with cathepsin D level (r = −0.17; P < 0.01), which had been determined previously in adjacent sections (17).

No significant difference in ERβ level attributable to menopausal and hormonal status was observed on this limited number of informed patients, contrasting with the ERα regulation observed in our study on 151 patients (17).

The variations in the levels of the ERs (α and β) during early stages of mammary carcinogenesis were studied by immunohistochemistry in epithelial mammary structures of increasing risk of developing invasive breast cancer, from simple BBD without cell proliferation to high-grade CIS. The percentage of ERα-positive cells is generally considered to be low (10–20%) in normal resting mammary glands (16, 17, 25, 26) and to increase in proliferative BBD (26), particularly when associated with atypia (17), and in low-grade DCIS. This increase has suggested an increased receptivity to estrogens in these tissues, contributing to their increased risk of tumorigenesis (27).

In this study, we show a marked decrease of ERβ-positive cells, statistically significant in the stage of proliferative breast disease without atypia and anticipating the ERα increase observed in BBDs with atypia. The ERβ level was further decreased in DCIS, mostly in high-nuclear-grade DCIS (median value, 3%). In proliferative BBD with atypia, there was additional increased ERα expression (17), resulting in a large decrease in the ERβ/ERα ratio. Therefore, in the histologically normal resting mammary gland, ERβ-positive cells (median, 85%) greatly exceed ERα-positive cells (median, 4%), whereas the ERα level progressively reached the ERβ level in BBD with atypia and in CIS. Ki67 staining, a marker of cell proliferation and, to a lesser extent, cathepsin D staining, was inversely correlated with the ERβ level.

Concerning mammary tumorigenesis, the decreased ERβ protein level that we observed in human preinvasive mammary lesions is in line with the decreased ERβ RNA level in invasive breast cancer tissues compared with the adjacent normal mammary gland (12) and with studies on mice mammary glands using the same antibodies (16). These results also support the hypothesis that ERβ might generally inhibit the mitogenic activity of estrogens mediated by ERα, as proposed from ERβ gene knock-out experiments showing increased Ki67 staining and signs of hyperplasia in mouse endometrium and prostate (9, 23, 24) and from ERβ RNA assays in other human estrogen-responsive cancers (28, 29).

The mechanism of the putative effect of ERβ on cell proliferation is unknown. Functional studies in cancer cell lines have shown differences in stimulating the transcriptional activating function (AF1) of the receptor (10) and activator protein (AP1) cross talk (11); however, the master genes involved in the mitogenic activity of estrogens are still being debated. Moreover, the ERβ expressed in stromal and vascular cells close to premalignant epithelial cells might also play a paracrine role in controlling cell proliferation (30). This study shows in preinvasive lesions a correlation between a decreased ERβ level and increased risk of breast cancer associated with these lesions. This decrease, as well as the increased Ki67 staining, might help to detect lesions with increasing risk. Long-term clinical follow-up of these patients could clarify this issue. Because the 503 IgY antibodies do not discriminate between the different ERβ isoforms (31, 32), it is possible that the proportion of these forms also modulates ERα function differently in mammary tumorigenesis. The mechanism of the ERβ down-regulation is probably to be sought within epithelial mammary cells because the level of ERβ in normal mammary glands (contrasting with that of ERα) was not altered by the varying level in the adjacent lesions.

We conclude that the expression of ERβ markedly decreases in the early stages of mammary carcinogenesis. The mechanism and role of this decrease in carcinogenesis are unknown. However, these results are consistent with a preventive effect of ERβ against the mitogenic effect of estrogens in human premalignant lesions and in line with the results of βERKO and αERKO in mice, indicating a stimulatory role of ERα and an inhibitory effect of ERβ in the proliferation of different estrogen-responsive tissues. Better knowledge of the ERα and ERβ content in each preinvasive lesion should help in the future to rationalize the use of preventive therapy with antiestrogen (33) adapted to each individual patient.

Fig. 1.

Specificity of immunohistochemical staining of ERβ. Adjacent sections from a normal lobule were treated with the chicken polyclonal ERβ 503 IgY alone (a), or after incubation with a 90-fold excess of pure ERβ (b) or with nonspecific IgY (c). Representative ERβ staining (brown) in normal duct (e), in nonproliferative BBD (cyst; f), in proliferative BBD without atypia (ductal hyperplasia; i), and in intermediate-grade DCIS (h). Adjacent sections of representative samples of normal duct (d and e) and of intermediate-grade DCIS (g and h) were stained with anti-ERβ antibody (chicken polyclonal ERβ 503 IgY, e and h) or with anti-ERα 1D5 antibody (d and g). Counterstaining (blue) was with hematoxylin. Bar, 50 μm.

Fig. 1.

Specificity of immunohistochemical staining of ERβ. Adjacent sections from a normal lobule were treated with the chicken polyclonal ERβ 503 IgY alone (a), or after incubation with a 90-fold excess of pure ERβ (b) or with nonspecific IgY (c). Representative ERβ staining (brown) in normal duct (e), in nonproliferative BBD (cyst; f), in proliferative BBD without atypia (ductal hyperplasia; i), and in intermediate-grade DCIS (h). Adjacent sections of representative samples of normal duct (d and e) and of intermediate-grade DCIS (g and h) were stained with anti-ERβ antibody (chicken polyclonal ERβ 503 IgY, e and h) or with anti-ERα 1D5 antibody (d and g). Counterstaining (blue) was with hematoxylin. Bar, 50 μm.

Close modal
Fig. 2.

Distribution of ERβ-positive epithelial cells. The percentage of stained nuclei is represented in normal glands (at the periphery of lesions), nonproliferative BBD, proliferative BBD without atypia, proliferative BBD with atypia, high-grade DCIS, and in other CISs. The number of different samples is in parentheses. Horizontal lines, median value; P according to nonparametric Kruskal-Wallis test. ∗, P < 0.0001 versus normal gland group. •, sclerosing adenosis, atypical lobular hyperplasia, and LCIS., intermediate-grade DCIS.

Fig. 2.

Distribution of ERβ-positive epithelial cells. The percentage of stained nuclei is represented in normal glands (at the periphery of lesions), nonproliferative BBD, proliferative BBD without atypia, proliferative BBD with atypia, high-grade DCIS, and in other CISs. The number of different samples is in parentheses. Horizontal lines, median value; P according to nonparametric Kruskal-Wallis test. ∗, P < 0.0001 versus normal gland group. •, sclerosing adenosis, atypical lobular hyperplasia, and LCIS., intermediate-grade DCIS.

Close modal
Fig. 3.

Comparison of ERβ level (expressed by the percentage of stained nuclei) between normal adjacent glands (a) and lesions (b). The number of different samples is in parentheses. Horizontal line, median value. The P between A and B according to the paired Wilcoxon test is significant (P < 0.007) for all lesions except in nonproliferative BBD. •, sclerosing adenosis, atypical lobular hyperplasia, and LCIS., intermediate-grade DCIS.

Fig. 3.

Comparison of ERβ level (expressed by the percentage of stained nuclei) between normal adjacent glands (a) and lesions (b). The number of different samples is in parentheses. Horizontal line, median value. The P between A and B according to the paired Wilcoxon test is significant (P < 0.007) for all lesions except in nonproliferative BBD. •, sclerosing adenosis, atypical lobular hyperplasia, and LCIS., intermediate-grade DCIS.

Close modal
Fig. 4.

Comparison between ERα level (α) and ERβ level (β), expressed as the percentage of stained nuclei, in normal glands (at the periphery of lesions) and the indicated lesions as in Fig. 2.

Fig. 4.

Comparison between ERα level (α) and ERβ level (β), expressed as the percentage of stained nuclei, in normal glands (at the periphery of lesions) and the indicated lesions as in Fig. 2.

Close modal
Fig. 5.

Evolution of the ratio between the ERβ and the ERα level (ERβ + 1/ERα + 1), expressed as the percentage of stained nuclei and represented on a log scale, in normal glands and the indicated lesions as in Fig. 2. ∗, the decreased ratio in proliferative lesions is highly significant, according to nonparametric Kruskal-Wallis test when compared with the normal gland group (P < 0.0001) or the nonproliferative BBD stage (P < 0.003).

Fig. 5.

Evolution of the ratio between the ERβ and the ERα level (ERβ + 1/ERα + 1), expressed as the percentage of stained nuclei and represented on a log scale, in normal glands and the indicated lesions as in Fig. 2. ∗, the decreased ratio in proliferative lesions is highly significant, according to nonparametric Kruskal-Wallis test when compared with the normal gland group (P < 0.0001) or the nonproliferative BBD stage (P < 0.003).

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by the Centre Hospitalier Universitaire of Montpellier, a grant from the French Ministry of Health (to H. R.), the Fondation pour la Recherche Médicale (to P. R.), the Swedish Cancer Fund (to J. A. G.), and the Institut National de la Santé et de la Recherche Médicale.

3

The abbreviations used are: CIS, carcinoma in situ; BBD, benign breast disease; ER, estrogen receptor; DCIS, ductal carcinoma in situ; LCIS, lobular carcinoma in situ.

4

M. Warner, S. Mäkelä, S. Nilsson, and J-Å. Gustafsson, unpublished experiments.

Table 1

Patient characteristics and distribution of breast lesions (see “Materials and Methods”)

A. Patient characteristics
AgeHormonal statusa
Median, years 49 Premenopausal, n (%) 35 (39) 
Range, years 30–79  Follicular phase, n 23 
H50 years, n (%) 60 (51)  Luteal phase, n 
>50 years, n (%) 58 (49) Menopausal, n (%) 55 (61) 
A. Patient characteristics
AgeHormonal statusa
Median, years 49 Premenopausal, n (%) 35 (39) 
Range, years 30–79  Follicular phase, n 23 
H50 years, n (%) 60 (51)  Luteal phase, n 
>50 years, n (%) 58 (49) Menopausal, n (%) 55 (61) 
B. Distribution of breast lesions
Patients, n (%)Lesions, n (%)
Nonproliferative BBD 11 (9) 19 (15) 1 mild ductal hyperplasia 
   8 adenosis 
   6 cysts without apocrine metaplasia 
   3 cysts with apocrine metaplasia 
Proliferative BBD without atypia 36 (31) 39 (30) 30 moderate and florid ductal hyperplasia 
   8 sclerosing adenosis 
   1 intraductal multiple papilloma 
Proliferative BBD with atypia 13 (11) 13 (10) 12 atypical ductal hyperplasia (3)b 
   1 atypical lobular hyperplasia 
CIS 58 (49) 59 (45) 56 DCIS 
   35 high-grade DCIS (21)b 
   7 intermediate-grade DCIS (5)b 
   14 low-grade DCIS (9)b 
   3 LCIS (2)b 
Total 118 130  
B. Distribution of breast lesions
Patients, n (%)Lesions, n (%)
Nonproliferative BBD 11 (9) 19 (15) 1 mild ductal hyperplasia 
   8 adenosis 
   6 cysts without apocrine metaplasia 
   3 cysts with apocrine metaplasia 
Proliferative BBD without atypia 36 (31) 39 (30) 30 moderate and florid ductal hyperplasia 
   8 sclerosing adenosis 
   1 intraductal multiple papilloma 
Proliferative BBD with atypia 13 (11) 13 (10) 12 atypical ductal hyperplasia (3)b 
   1 atypical lobular hyperplasia 
CIS 58 (49) 59 (45) 56 DCIS 
   35 high-grade DCIS (21)b 
   7 intermediate-grade DCIS (5)b 
   14 low-grade DCIS (9)b 
   3 LCIS (2)b 
Total 118 130  
a

In 28 patients, the hormonal and menopausal status could not be defined.

b

Number of lesions at the periphery of invasive carcinomas.

Table 2

Distribution of Ki67- and ERβ-positive epithelial cells in 61 lesions from 54 patients

Normal glandNonproliferative BBDProliferative BBD without atypiaProliferative BBD with atypiaNon-high-grade CISHigh-grade DCIS
n              a 54 11 13 10 20 
Ki67       
 Median (%)b 1.5 3.5 16 6.1 17.3 
 Range (%)b 0–19.3 0–12.5 0–13.4 0–40.8 1.8–38.8 2.5–70.3 
ERβ       
 Median (%)b 84.7 75.4 18.5 13.5 33.4 
Normal glandNonproliferative BBDProliferative BBD without atypiaProliferative BBD with atypiaNon-high-grade CISHigh-grade DCIS
n              a 54 11 13 10 20 
Ki67       
 Median (%)b 1.5 3.5 16 6.1 17.3 
 Range (%)b 0–19.3 0–12.5 0–13.4 0–40.8 1.8–38.8 2.5–70.3 
ERβ       
 Median (%)b 84.7 75.4 18.5 13.5 33.4 
a

Number of structures.

b

Percentage of stained nuclei.

We thank Drs. Daniel Dupeigne, Pierre-Ludovic Giacalone, Christine Pignodel, François Laffargue, Pierre Marès, Christiane Marty Double, and Thierry Maudelonde for their valuable contributions in providing tissue samples. We also thank Jean-Yves Cance for preparing the figures, and Nadia Kerdjadj and Michèle Troc for secretarial assistance.

1
Vogelstein B., Kinzler K. W. The multistep nature of cancer.
Trends Genet.
,
9
:
138
-141,  
1993
.
2
Dupont W. D., Page D. L. Risk factors for breast cancer in women with proliferative breast disease.
N. Engl. J. Med.
,
312
:
146
-151,  
1985
.
3
O’Connell P., Pekkel V., Fuqua S. A., Osborne C. K., Clark G. M., Allred D. C. Analysis of loss of heterozygosity in 399 premalignant breast lesions at 15 genetic loci.
J. Natl. Cancer Inst. (Bethesda)
,
90
:
697
-703,  
1998
.
4
Laidlaw I. J., Clarke R. B., Howell A., Owen A. W. M. C., Potten C. S., Anderson E. The proliferation of normal human breast tissue implanted into athymic nude mice is stimulated by estrogen but not progesterone.
Endocrinology
,
136
:
164
-171,  
1995
.
5
Kuiper G. G., Enmark E., Pelto-Huikko M., Nilsson S., Gustafsson J. A. Cloning of a novel receptor expressed in rat prostate and ovary.
Proc. Natl. Acad. Sci. USA
,
93
:
5925
-5930,  
1996
.
6
Enmark E., Pelto-Huikko M., Grandien K., Lagercrantz S., Lagercrantz J., Fried G., Nordenskjold M., Gustafsson J. A. Human estrogen receptor β-gene structure, chromosomal localization, and expression pattern.
J. Clin. Endocrinol. Metab.
,
82
:
4258
-4265,  
1997
.
7
Taylor A. H., Al-Azzawi F. Immunolocalisation of oestrogen receptor β in human tissues.
J. Mol. Endocrinol.
,
24
:
145
-155,  
2000
.
8
Korach K. S. Insights from the study of animals lacking functional estrogen receptor.
Science (Washington DC)
,
266
:
1524
-1527,  
1994
.
9
Krege J. H., Hodgin J. B., Couse J. F., Enmark E., Warner M., Mahler J. F., Sar M., Korach K. S., Gustafsson J. A., Smithies O. Generation and reproductive phenotypes of mice lacking estrogen receptor β.
Proc. Natl. Acad. Sci. USA
,
95
:
15677
-15682,  
1998
.
10
Cowley S. M., Parker M. G. A comparison of transcriptional activation by ERα and ERβ.
J. Steroid Biochem.
,
69
:
165
-175,  
1999
.
11
Paech K., Webb P., Kuiper G. G., Nilsson S., Gustafsson J. A., Kushner P. J., Scanlan T. S. Differential ligand activation of estrogen receptors ERα and ERβ at AP1 sites.
Science (Washington DC)
,
277
:
1508
-1510,  
1997
.
12
Leygue E., Dotzlaw H., Watson P. H., Murphy L. C. Altered estrogen receptor α and β messenger RNA expression during human breast tumorigenesis.
Cancer Res.
,
58
:
3197
-3201,  
1998
.
13
Jarvinen T. A., Pelto-Huikko M., Holli K., Isola J. Estrogen receptor β is coexpressed with ERα and PR and associated with nodal status, grade, and proliferation rate in breast cancer.
Am. J. Pathol.
,
156
:
29
-35,  
2000
.
14
Speirs V., Parkes A. T., Kerin M. J., Walton D. S., Carleton P. J., Fox J. N., Atkin S. L. Coexpression of estrogen receptor α and β: poor prognostic factors in human breast cancer?.
Cancer Res.
,
59
:
525
-528,  
1999
.
15
Greene G. L., Nolan C., Engler J. P., Jensen E. V. Monoclonal antibodies to human estrogen receptor.
Proc. Natl. Acad. Sci. USA
,
77
:
5115
-5119,  
1980
.
16
Saji S., Jensen E. V., Nilsson S., Rylander T., Warner M., Gustafsson J. A. Estrogen receptors α and β in the rodent mammary gland.
Proc. Natl. Acad. Sci. USA
,
97
:
337
-342,  
2000
.
17
Roger P., Daures J. P., Maudelonde T., Pignodel C., Gleizes M., Chapelle J., Marty-Double C., Balder P., Mares P., Laffargue F., Rochefort H. Dissociated overexpression of cathepsin-D and estrogen receptor α in preinvasive mammary tumors.
Hum. Pathol.
,
31
:
593
-600,  
2000
.
18
Silverstein M. J., Poller D. N., Waisman J. R., Colburn W. J., Barth A., Gierson E. D., Lewinsky B., Gamagami P., Slamon D. J. Prognostic classification of breast ductal carcinoma in situ.
Lancet
,
345
:
1154
-1157,  
1995
.
19
Consensus conference on the classification of ductal carcinoma in situ.
Hum. Pathol.
,
28
:
1221
-1225,  
1997
.
20
Al Saati T., Clamens S., Cohen-Knafo E., Faye J. C., Prats H., Coindre J. M., Wafflart J., Caverivière P., Bayard F., Delsol G. Production of monoclonal antibodies to human estrogen-receptor protein (ER) using recombinant ER (RER).
Int. J. Cancer
,
55
:
651
-654,  
1993
.
21
Norton A. J., Jordan S., Yeomans P. Brief, high-temperature heat denaturation (pressure cooking): an effective method of antigen retrieval for routinely processed tissues.
J. Pathol.
,
173
:
371
-379,  
1994
.
22
Key G., Becker M. H., Baron B., Duchrow M., Schluter C., Flad H. D., Gerdes J. New Ki-67-equivalent murine monoclonal antibodies (MIB 1–3) generated against bacterially expressed parts of the Ki-67 cDNA containing three 62 base pair repetitive elements encoding for the epitope.
Lab. Investig.
,
68
:
629
-636,  
1993
.
23
Gustafsson J. A. Estrogen receptor β—a new dimension in estrogen mechanism of action.
J. Endocrinol.
,
163
:
379
-383,  
1999
.
24
Weihua Z., Saji S., Makinen S., Cheng G., Jensen E. V., Warner M., Gustafsson J. A. Estrogen receptor (ER) β, a modulator of ERα in the uterus.
Proc. Natl. Acad. Sci. USA
,
97
:
5936
-5941,  
2000
.
25
Clarke R. B., Howell A., Potten C. S., Anderson E. Dissociation between steroid receptor expression and cell proliferation in the human breast.
Cancer Res.
,
57
:
4987
-4991,  
1997
.
26
Khan S. A., Rogers M. A., Khurana K. K., Meguid M. M., Numann P. J. Estrogen receptor expression in benign breast epithelium and breast cancer risk.
J. Natl. Cancer Inst. (Bethesda)
,
90
:
37
-42,  
1998
.
27
Shoker B. S., Jarvis C., Clarke R. B., Anderson E., Hewlett J., Davies M. P., Sibson D. R., Sloane J. P. Estrogen receptor-positive proliferating cells in the normal and precancerous breast.
Am. J. Pathol.
,
155
:
1811
-1815,  
1999
.
28
Pujol P., Rey J. M., Nirde P., Roger P., Gastaldi M., Laffargue F., Rochefort H., Maudelonde T. Differential expression of estrogen receptor-α and -β messenger RNAs as a potential marker of ovarian carcinogenesis.
Cancer Res.
,
58
:
5367
-5373,  
1998
.
29
Foley E. F., Jazaeri A. A., Shupnik M. A., Jazaeri O., Rice L. W. Selective loss of estrogen receptor a in malignant human colon.
Cancer Res.
,
60
:
247
-248,  
2000
.
30
Cunha G. R., Young P., Hom Y. K., Cooke P. S., Taylor J. A., Lubahn D. B. Elucidation of a role for stromal steroid hormone receptors in mammary gland growth and development using tissue recombinants.
J. Mammary Gland Biol. Neoplasia
,
2
:
393
-402,  
1997
.
31
Hanstein B., Liu H., Yancisin M. C., Brown M. Functional analysis of a novel estrogen receptor-β isoform.
Mol. Endocrinol.
,
13
:
129
-137,  
1999
.
32
Ogawa S., Inoue S., Watanabe T., Orimo A., Hosoi T., Ouchi Y., Muramatsu M. Molecular cloning and characterization of human estrogen receptor betacx: a potential inhibitor of estrogen action in human.
Nucleic Acids Res.
,
26
:
3505
-3512,  
1998
.
33
Fisher B., Costantino J. P., Wickerham D. L., Redmond C. K., Kavanah M., Cronin W. M., Vogel V., Robidoux A., Dimitrov N., Atkins J., Daly M., Wieand S., Tan-Chiu E., Forl L., Wolmark N. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 study.
J. Natl. Cancer Inst. (Bethesda)
,
90
:
1371
-1388,  
1998
.