BIOCHEMICAL
Vol. 166, No. 2, 1990 January 30, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 601-607
ESTROGEN RECEPTOR GENE AMPLIFICATION ESTROGEN RECEPTOR-POSITIVE HUMAN
IS FOUND IN SOME BREAST TUMORS
Monica Nembrot*, Brenda Quintana** and JoseMordoh*# *Institute de Investigaciones Bioquimicas “Fundacion Campomar”, Patricias Argentinas 435, 1405 Buenos Aires, Argentina and ** Hospital Ramos Mejia, Gral. Urquiza 609, 1221Buenos Aires, Argentina Received
November
20,
1989
The genomic organization of the estrogen receptor (ER) gene has been analyzed in 21 primary human breast cancers and 1 axillary metastasis.No evidence of rearrangements of the ER gene was found in the analyzed tumors. In 6/14 ERpositive tumors a certain degree of amplification of the ER gene, ranging from 1.6 to 3-fold, was detected. No correlation was observed between the level of gene amplification and the amount of ER in the tumors. In the 8 ER-negative tumors analyzed no amplification could be detected. It is concluded that ER gene amplification may be one of the mechanismsunderlying the increased ER expression in somebreast tumors. 0 1990Academic Press,Inc.
The female hormone 17-P-estradiol (E2) regulates the tissue-specific genetic expression in target cells through its interaction with specific estrogen receptors (ER) (1). Several cDNAs of the human ER have been recently cloned (2, 3), providing powerful tools to analyze the structure and expression of the ER gene. The ER expression in breast, one of the E, target organs, is yet poorly understood either in normal or pathological conditions. Whereas in adult normal breast only a small number of ER-positive cells are found (4), more than 60% of human breast cancers display significative amounts of ER (5), suggestinga role of the ER in the genesisor growth of human breast cancer. Recent work has shown that in some pathological conditions the increased expression of a given protein may be associated with genetic alterations. As an example, the correlation between the increased c-myc expression and the 814 chromosome translocation found in Burkitt’s lymphoma (6) may be quoted. In 2530% of human breast cancers an amplification of the Her-2/neu proto-oncogene and a concomitant overexpression of the encoded ~185 protein has been demonstrated (7). Furthermore, the presence of aneuploidy in more than 50% of human breast cancers has been repeatedly shown, predominantly in ER-negative tumors (8-10). The purpose of the work reported in this paper was to analyze if the increased ER
#To whom requests for reprints should be addressed. Abbreviations: ER, estrogen receptor; E2, estradiol; kb, kilobase; PgR, progesterone receptor. 0006-291X/90
601
$1.50
Copyright 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.
Vol.
166,
No.
expression
2, 1990
in ER-positive
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
breast tumors or the nil ER expression
COMMUNICATIONS
in ER-negative
tumors may be associated with alterations in the ER gene. MATERIALS
AND METHODS
Patients. Breast primary tumors and an axillary metastasis were obtained from 22 previously untreated patients whose clinical stage was determined by the TNM classification. Tissue preservation. The breast tumors were obtained from mastectomies or tumorectomies, cleansed immediately from adipose and surrounding normal tissues and frozen in liquid nitrogen until their analysis. Samples of normal endometrium were obtained from three patients submitted to histerectomy due to uterine myoma. The endometria were scrapped and frozen in liquid nitrogen. The placenta was obtained from a woman after normal delivery, it was washed with PBS to remove blood and kept at -70” C. Cell lines. The ER-positive human breast cancer cell line MCF-7 (11) and the ERnegative human breast cancer cell line MDA-231 (12) were grown as previously described (13). The cultures were incubated at 37°C in a humid incubator under an air-CO2 (95:5) atmosphere. Analvsis of DNA. High molecular weight genomic DNAs were prepared as previously described (14). 40 pg of DNA of each sample were digested overnight at 37°C with 200 units of EcoR I (Bethesda Research Laboratories). After digestion, electrophoresis was performed in 0.8% agarose gels and the DNAs were transferred to nylon Z-Probe membranes (Bio-Rad) in alkaline solution (15). x DNA digested with Hind III was used as size marker. Plasmids. The recombinant plasmid HE0 containing an 1.8 kilobase (kb) insert corresponding to the complete human ER cDNA and the plasmid HE14 containing an 1.2 kb cDNA insert corresponding to the hormone-binding site of the human ER (16) were generously provided by Dr. Pierre Chambon. A plasmid containing an 1.3 kb EcoR I-BamH I fragment of the coding region of mouse 28s ribosomal RNA gene (rRNA) (17) was provided by Dr. Chantal Escot. Plasmid pFN4H 0.95 contains a 0.95 kb Hind III fragment of the human fibronectin gene (18) and was provided by Dr. Albert0 Kornblihtt. The inserts were purified by electroelution (19) ar$ passage through Elutip-D (Schleicher and Schuell) and were mck-translated with [- P] dCJTP to a specific activity of 1-2x lo8 cpm/pg DNA. Hybridization. After DNA transfer, the membranes were washed in 6 x SSC and baked for 2 hr at 80” C. The filters were prehybridized overnight at 65 ‘C in a solution containing 6 x SSPE, 0.1% SDS, 100 &ml salmon sperm DNA and 10 x Denhardt’s solution. Hybridization was erformed during 48 hr at 65°C in sealed plastic bags containing 6 x SSPE, 0.1% g DS, 100 &ml salmon sperm DNA, 10 x Denhardt’s solution, 10% dextran sulfate and the radioactive probe (1 x lo6 cpm/ml). After hybridization, the membranes were washed in high stringency conditions: 30 min at room temperature in 2 x SSC, 0.1% SDS, and 60 min at 65 “C in 0.1 x SSC, 0.1% SDS. After washing, the membranes were autoradiographed with Kodak X-OMAT films at -70” C before development. The membranes were deh bridized by heating twice at 95°C (20 min each) in a solution containing 0.1 x HSC, 0.5% SDS, and were rehybridized as described above. Densitometrv. The quantification of the autoradiographic signals was performed using a densitometer (Helena Laboratories Corp., USA). The amount of the ERDNA present was calculated after scanning of the 3.6 kb band. As controls, two unrelated enes were also measured: the 6.2 kb band of the multiple-copy rRNA gene and t Pie 4.5 kb band (18) of the single copy fibronectin gene. The ratio between the ER and ribosomal bands and/or ER and fibronectin bands present in placenta was taken as the unity. 602
Vol.
166,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Hormone receptor determinations. The ER and progesterone receptor (PgR) were determined by the DCC technique (20) on samples frozen in liquid nitrogen. RESULTS Normal
eenomic pattern
of the ER eene. Our first experiments
were
directed to
determine the normal pattern of the ER gene in human tissues, and if alterations that pattern could be found in tissues with widely divergent ER expression placenta
and endometrium.
DNAs
isolated from ER-positive
(MCF-7)
of
such as and ER-
negative (MDA) human breast cancer cell lines were also analyzed. Hybridizations were performed with the HE0 plasmid that codes for the entire ER cDNA sequence. In every case the obtained pattern was identical, eight bands of 8.5 kb, 7.6 kb, 6.9 kb, 5.0 kb, 3.6 kb, 3.1 kb, 2.7 kb and 1.7 kb being observed (Fig. 1A). These results are in accordance
with those reported
dehybridized
and rehybridized
for the MCF-7 with the rRNA
cell line (2). The filters were later (Fig. 1C) and fibronectin
(Fig. 1D)
probes. Hybridizations with these two probes, that detect a multiple-copy gene and a single-copy gene respectively, were used to control variations in the amount of the ER gene due to technical reasons. When the ratios ER/rRNA
were analyzed, a ratio
similar to placenta was found for the different samples. The same situation was found when the ratios ER/fibronectin DNA
were analyzed, with the only exception of the MCF-7
(ratio = 1.5). With this possible exception, it may be concluded that the ER
gene is not amplified in any of these tissues or cells. With the purpose of identifying domains
of the ER, hybridizations
contains
the coding sequences
DNA-binding
the genomic bands coding for the different
were
performed
with the HE14 plasmid that
for the hormone-binding
domain but not for the
domain of the ER. In Fig. 1B it may be observed that only the 8.5 kb,
7.6 kb, 6.9 kb, 3.6 kb and 2.7 kb bands are detected. Genomic pattern of the ER eene in Drimarv
breast tumors.
A series of ER-negative
and ER-positive tumors were analyzed to detect any qualitative or quantitative changes in the ER gene. Some of the results obtained are illustrated in Fig. 2A, where a series of ER-positive
tumors were analyzed by Southern blots with the HE0
probe. In the same figure the results of hybridization
with the rRNA
and fibronectin
probes are shown (Figs. 2B and 2C respectively). When the ratio between the HE0 and rRNA bands for placenta was taken as the unity, the ratio for tumors #9, #14, #4, #7 and #6 were 3.0, 1.6, 1.0, 1.0 and 1.0, respectively. When the ratio between the HE0 and the fibronectin bands for placenta was taken as the unity, identical results were obtained. As a consequence, the first ratio was used for the analysis of every tumor. The relevant clinical characteristics of the patients studied, the ER gene amplification
and the biochemical
determinations
of the ER and PgR are indicated
in Table 1. The results obtained, shown in Table 1 and illustrated in Fig. 2, demonstrate that: 1) the ER genomic pattern was similar in both ER-negative and ER-positive
tumors;
2) a single copy of the ER-gene was found in every analyzed 603
Vol.
166, No. 2, 1990
BIOCHEMICAL
A kb 1 2 3 4 5
kbl
AND BIOPHYSICAL
B 2 34
RESEARCH COMMUNICATIONS
5 A
kb 1 2 3 4 5 6
6.5 * 7.6: 6.9 5.0 -
3.63.1 2.7 -
B 6.2, 4.5-
0
0
1
2
C 4.5 -D
Figure 1, ER geneanaivsisin human breastcancer ceil lines MCF-7 and MDA, placentaand normal endometrium.GenomicDNAs from thesehumantissueswere digestedwith EkoR I, electrophoresed,blotted and hybridized with plasmidsHE0 (A), HE14 (B), rRNA (C) and fibronectin (D). 1: placenta;2: MCF-7; 3: MDA; 4: endometrium#l; 5: endometrium#2. FiPure 2. ER Peneanalvsisin human breast cancer. DNA samplesfrom primary breast tumors were restricted with EcoR I, electrophoresed,blotted and after hybridizing with probe HEO (A), they were rehybridizedwith probe for rRNA (B) and fibronectin (C). Five representativesamplesof the tumorsanalyzedare shown. 1: #9; 2: #14; 3: #4; 4: #7; 5: #6; 6: placenta.
ER-negative tumor (8/8); 3) in some ER-positive tumors (6/14) an amplification of the ER-gene was detected. Even when present, the detected amplification was relatively low (up to 3- fold) and was not directly related to the ER content. DISCUSSION The discrimination of human breast cancers into ER-positive and ERnegative has great prognostic value. Thus, about 60% of patients with ER-positive tumors respond to hormonal treatments as compared to only 510% of patients with ER-negative tumors (21). It is therefore evident that the ability of human breast cancer to express ER is associatedwith a different biological behavior. The reasons underlying that increased expression are not clearly understood. A priori, several possibilities may be envisaged: a) an increased gene dosage of the ER gene; b) an increased transcriptional activity of that gene; c) modifications at the translational or post-translational level. In this paper we have provided evidence strongly suggesting that a higher gene dosage may play a role in the increased amount of ER found in some ER-positive tumors. This is based in the fact that in 6/14 ER-positive tumors a 604
Vol.
BIOCHEMICAL
166, No. 2, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
TableI Correlation betweenER content andER genecopynumber Patient
Age
Clinical Stage
:
5:
:I
445 397
241 0
1.6
ii 5
:t 57;
&I IIIb
308 275 200
207 60 71
kil:o 1.0
IIa I
141 187
1:s
::i
:
ER PgR (fmoles/mg prot)
ER (no of copies)
;
2
IIb
103
241 348
:::
11 10
2:
Ihb
2; 32
169 10
E+
12 13 14
5: 68 35
i;
14 12 13
0
IIIa II
208
E 1:6
15 16
61 ::
IIIb II
::
i
1.0 1.0
17 18 19
i;’
I IIIa
0 0
0 i
::“o 1.0
Z* 22
2: 83
rr :::
0i 0
0 ii
::i 1.0
The clinical characteristics of the patients studied and the ER and PgR values are indicated. After extraction and digestion with EcoR I, DNA was electrophoresed, blotted and the filters were hybridized sequentially with the HE0 and rRNA probes as described under Methods. For estimating the ER gene copy number, the 3.6 kb band of the ER-gene and the 6.2 kb band detected with the rRNA probe were densitometrically measured. The ratio ER/rRNA obtained for placenta was taken as 1. * Axillary metastasis.
certain degree of amplification of the ER gene is found, asopposed to a normal level in the 8 ER-negative tumors analyzed. Although the level of amplification detected was relatively low (1.6 to 3.0-fold), it should be taken into account that these are minimal values since there is a significative contribution of normal fibroblasts and lymphoid cells in human breast cancer tissues. However, the possibility that the increased dosage of the ER gene is due to a partial or complete duplication of chromosome 6 should also be considered. Although the number of casesanalyzed are too few to draw any definite conclusions, no correlation was found between the level of gene amplification and ER expression. The mechanism of ER expression in breast cancer appears to be quite complex. Thus, Ballare et al. (22) have recently found that in ER-positive tumors about 70% of the DNA-synthesizing cells are ERnegative, suggestingthat ER are not present in the most primitive breast cancer cells, but are acquired as the differentiation proceeds. The same authors have found that the stability of the ER differs among ER-positive tumors, pointing to possible differences either in the ER molecule itself or in the cellular processingmechanisms. 605
Vol. 166, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Recently, several species of ER mRNA have been found in human breast cancers, suggesting additional mechanisms of variation in the ER content (23). It is noteworthy that no changesin the EcoR I restriction pattern of the ER gene have been detected in any of the analyzed tumors. Other authors arrived to the same results with enzymes like BamH I, Hind III, Sst I and Pst I, although with Pvu II two polymorphic alleles in the hormone binding domain of the ER gene were found (24). These results support the idea that there are no extensive chromosomal rearrangements or deletions in the ER gene, even in ER-negative tumors, where the incidence of chromosomal changesis considerably larger (8-10). ACKNOWLEDGMENTS This work has been supported by Grants from the Consejo National de Investigaciones Cientificas y Tecnicas (CONICET), Argentina and the Comision de Investigaciones Cientificas y Tecnicas de la Provincia de Buenos Aires, Argentina. J. M. is a Career Investigator and M. N. a Fellow of the former Institution. We are grateful to Dr. C. Levy (Hospital Ramos Mejia, Buenos Aires, Argentina) and Dr. C. Garbovesky (Hospital Municipal de Oncologia, Buenos Aires, Argentina) for the provision of the tumors. REFERENCES 1. Anderson J. N. (1984) in Biological Regulation and Development, Goldenberg, R. F. and Yamamoto, K. R., eds. (Plenum, New York), Vol. 3B, 169-212. 2. Walter P., Green S., Greene G., Krust A., Bornet J-M, Jeltsch J-M, Staub A., Jensen E., Scrace G., Waterfield M. and Chambon P. (1985) Proc. Natl. Acad. Sci. (USA) 82,7889-7893. 3. Green S., Walter P., Kumar V., Krust A., Bornet J-M., Argos P. and Chambon P. (1986) Nature 320,134-139. 4. Petersen 0. W., Hoyer P. E. and van Deurs B. 1987) Cancer Res. 47,5748-5751. 5. Jensen E. V., Polley T. Z., Smith S., Block G. Ii ., Ferguson D. J. and DeSombre E. R. (1975) in Estrogen Receptors in Human Breast Cancer, McGuire W. L., Carbone P. P. and Vollmer E. P., eds. (Raven Press,New York) Chapter 5,37-56. 6. Nishikura K., Ar-Rushdi A., Erikson J., Watt R., Rovera G. and Croce C. M. (1983) Proc. Natl. Acad. Sci. (USA) 80,4822-4826. 7. Slamon D. J., Godolphin W., Jones L. A., Holt J. A., Wong S. G., Keith D. E., Levin W. J., Stuart S. G., Udove J., Urlich A. and Press M. F. (1989) Science 244, 707-712. 8. Dressler L. G., Seamer L. C., Owens M. A., Clark G. M. and McGuire W. L. (1988) Cancer 61,420-427. 9. Hedley D. W., Rugg C. A., Ng A. B. P. and Taylor I. W. (1984) Cancer Res. 44, 5395-5398. 10. Devilee P., Thierry R. F., Kievits T., Kolluri R., Hopman A. H. N., Willard H., Pearson P. L. and CornelisseC. J. (1988) Cancer Res. 48,5825-5830. 11. Soule H. D., Vazquez J., Long A., Albert S. and Brennan M. A. (1973) J. Natl. Cancer Inst. 51,1409-1416. 12. Cailleau R., Young R., Olive M., Reeves W. J. Jr. (1974) J. Natl. Cancer Inst. 53, 661-674. 13. Resnicoff M., Medrano E. E., Podhajhcer 0. L., Bravo A. I., Bover L. and Mordoh J. (1987) Proc. Natl. Acad. Sci. (USA) 84,7295-7299. 14. Davis R. W., Thomas M., Cameron J., St. John T. P., Scherer S. and Padgett R. A. (1980) Meth. Enzymol. 65,404-411. 15. Reed K. C. and Mann V. A. (1985) Nucleic Acid Res. 13,7207-7221. 16. Kumar V., Green S., Staub A. and Chambon P. (1986) Embo J. 5,2231-2236. 17. Bowman L. H., Rabin B. and SchlessingerD. (1981) Nucleic Acid Res. 9, 49514967. 606
Vol.
166,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
18. Gutman A. and Kornblihtt A. (1987) Proc. Natl. Acad. Sci. USA) 84,7179-7182. 19. McDonell M. W., Simon M. N. and Studier F. W. (1977) 5 . Mol. Biol. 110, 119-
146. 20. McGuire 989.
W. L. and De La Garza M. (1973) J. Clin. Endocrinol.
Metab. 37, 986-
21. Edwards D. P., Chamness G. C. and McGuire W. L. (1979) Biochem. Biophys. Acta 560.457-486. 22. Ball&e C., Bravo A. I., Laucella S., Sorin I., Cerdeiro R., Loza J., Sousa Martinez F., Guman N. and Mordoh J. (1989) Cancer 64,842-848. 23. Murphy L. C. and Dotzlaw H. (1989) Mol. Endocr. 3,687-693. 24. Hill S. M., Fuqua S. A. W., Chamness G. C., Greene G. L. and McGuire W. L. (1989) Cancer Res. 49,145148.
607
Vol. 166, No. 2, 1990 January 30, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 601-607
ESTROGEN RECEPTOR GENE AMPLIFICATION ESTROGEN RECEPTOR-POSITIVE HUMAN
IS FOUND IN SOME BREAST TUMORS
Monica Nembrot*, Brenda Quintana** and JoseMordoh*# *Institute de Investigaciones Bioquimicas “Fundacion Campomar”, Patricias Argentinas 435, 1405 Buenos Aires, Argentina and ** Hospital Ramos Mejia, Gral. Urquiza 609, 1221Buenos Aires, Argentina Received
November
20,
1989
The genomic organization of the estrogen receptor (ER) gene has been analyzed in 21 primary human breast cancers and 1 axillary metastasis.No evidence of rearrangements of the ER gene was found in the analyzed tumors. In 6/14 ERpositive tumors a certain degree of amplification of the ER gene, ranging from 1.6 to 3-fold, was detected. No correlation was observed between the level of gene amplification and the amount of ER in the tumors. In the 8 ER-negative tumors analyzed no amplification could be detected. It is concluded that ER gene amplification may be one of the mechanismsunderlying the increased ER expression in somebreast tumors. 0 1990Academic Press,Inc.
The female hormone 17-P-estradiol (E2) regulates the tissue-specific genetic expression in target cells through its interaction with specific estrogen receptors (ER) (1). Several cDNAs of the human ER have been recently cloned (2, 3), providing powerful tools to analyze the structure and expression of the ER gene. The ER expression in breast, one of the E, target organs, is yet poorly understood either in normal or pathological conditions. Whereas in adult normal breast only a small number of ER-positive cells are found (4), more than 60% of human breast cancers display significative amounts of ER (5), suggestinga role of the ER in the genesisor growth of human breast cancer. Recent work has shown that in some pathological conditions the increased expression of a given protein may be associated with genetic alterations. As an example, the correlation between the increased c-myc expression and the 814 chromosome translocation found in Burkitt’s lymphoma (6) may be quoted. In 2530% of human breast cancers an amplification of the Her-2/neu proto-oncogene and a concomitant overexpression of the encoded ~185 protein has been demonstrated (7). Furthermore, the presence of aneuploidy in more than 50% of human breast cancers has been repeatedly shown, predominantly in ER-negative tumors (8-10). The purpose of the work reported in this paper was to analyze if the increased ER
#To whom requests for reprints should be addressed. Abbreviations: ER, estrogen receptor; E2, estradiol; kb, kilobase; PgR, progesterone receptor. 0006-291X/90
601
$1.50
Copyright 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.
Vol.
166,
No.
expression
2, 1990
in ER-positive
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
breast tumors or the nil ER expression
COMMUNICATIONS
in ER-negative
tumors may be associated with alterations in the ER gene. MATERIALS
AND METHODS
Patients. Breast primary tumors and an axillary metastasis were obtained from 22 previously untreated patients whose clinical stage was determined by the TNM classification. Tissue preservation. The breast tumors were obtained from mastectomies or tumorectomies, cleansed immediately from adipose and surrounding normal tissues and frozen in liquid nitrogen until their analysis. Samples of normal endometrium were obtained from three patients submitted to histerectomy due to uterine myoma. The endometria were scrapped and frozen in liquid nitrogen. The placenta was obtained from a woman after normal delivery, it was washed with PBS to remove blood and kept at -70” C. Cell lines. The ER-positive human breast cancer cell line MCF-7 (11) and the ERnegative human breast cancer cell line MDA-231 (12) were grown as previously described (13). The cultures were incubated at 37°C in a humid incubator under an air-CO2 (95:5) atmosphere. Analvsis of DNA. High molecular weight genomic DNAs were prepared as previously described (14). 40 pg of DNA of each sample were digested overnight at 37°C with 200 units of EcoR I (Bethesda Research Laboratories). After digestion, electrophoresis was performed in 0.8% agarose gels and the DNAs were transferred to nylon Z-Probe membranes (Bio-Rad) in alkaline solution (15). x DNA digested with Hind III was used as size marker. Plasmids. The recombinant plasmid HE0 containing an 1.8 kilobase (kb) insert corresponding to the complete human ER cDNA and the plasmid HE14 containing an 1.2 kb cDNA insert corresponding to the hormone-binding site of the human ER (16) were generously provided by Dr. Pierre Chambon. A plasmid containing an 1.3 kb EcoR I-BamH I fragment of the coding region of mouse 28s ribosomal RNA gene (rRNA) (17) was provided by Dr. Chantal Escot. Plasmid pFN4H 0.95 contains a 0.95 kb Hind III fragment of the human fibronectin gene (18) and was provided by Dr. Albert0 Kornblihtt. The inserts were purified by electroelution (19) ar$ passage through Elutip-D (Schleicher and Schuell) and were mck-translated with [- P] dCJTP to a specific activity of 1-2x lo8 cpm/pg DNA. Hybridization. After DNA transfer, the membranes were washed in 6 x SSC and baked for 2 hr at 80” C. The filters were prehybridized overnight at 65 ‘C in a solution containing 6 x SSPE, 0.1% SDS, 100 &ml salmon sperm DNA and 10 x Denhardt’s solution. Hybridization was erformed during 48 hr at 65°C in sealed plastic bags containing 6 x SSPE, 0.1% g DS, 100 &ml salmon sperm DNA, 10 x Denhardt’s solution, 10% dextran sulfate and the radioactive probe (1 x lo6 cpm/ml). After hybridization, the membranes were washed in high stringency conditions: 30 min at room temperature in 2 x SSC, 0.1% SDS, and 60 min at 65 “C in 0.1 x SSC, 0.1% SDS. After washing, the membranes were autoradiographed with Kodak X-OMAT films at -70” C before development. The membranes were deh bridized by heating twice at 95°C (20 min each) in a solution containing 0.1 x HSC, 0.5% SDS, and were rehybridized as described above. Densitometrv. The quantification of the autoradiographic signals was performed using a densitometer (Helena Laboratories Corp., USA). The amount of the ERDNA present was calculated after scanning of the 3.6 kb band. As controls, two unrelated enes were also measured: the 6.2 kb band of the multiple-copy rRNA gene and t Pie 4.5 kb band (18) of the single copy fibronectin gene. The ratio between the ER and ribosomal bands and/or ER and fibronectin bands present in placenta was taken as the unity. 602
Vol.
166,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Hormone receptor determinations. The ER and progesterone receptor (PgR) were determined by the DCC technique (20) on samples frozen in liquid nitrogen. RESULTS Normal
eenomic pattern
of the ER eene. Our first experiments
were
directed to
determine the normal pattern of the ER gene in human tissues, and if alterations that pattern could be found in tissues with widely divergent ER expression placenta
and endometrium.
DNAs
isolated from ER-positive
(MCF-7)
of
such as and ER-
negative (MDA) human breast cancer cell lines were also analyzed. Hybridizations were performed with the HE0 plasmid that codes for the entire ER cDNA sequence. In every case the obtained pattern was identical, eight bands of 8.5 kb, 7.6 kb, 6.9 kb, 5.0 kb, 3.6 kb, 3.1 kb, 2.7 kb and 1.7 kb being observed (Fig. 1A). These results are in accordance
with those reported
dehybridized
and rehybridized
for the MCF-7 with the rRNA
cell line (2). The filters were later (Fig. 1C) and fibronectin
(Fig. 1D)
probes. Hybridizations with these two probes, that detect a multiple-copy gene and a single-copy gene respectively, were used to control variations in the amount of the ER gene due to technical reasons. When the ratios ER/rRNA
were analyzed, a ratio
similar to placenta was found for the different samples. The same situation was found when the ratios ER/fibronectin DNA
were analyzed, with the only exception of the MCF-7
(ratio = 1.5). With this possible exception, it may be concluded that the ER
gene is not amplified in any of these tissues or cells. With the purpose of identifying domains
of the ER, hybridizations
contains
the coding sequences
DNA-binding
the genomic bands coding for the different
were
performed
with the HE14 plasmid that
for the hormone-binding
domain but not for the
domain of the ER. In Fig. 1B it may be observed that only the 8.5 kb,
7.6 kb, 6.9 kb, 3.6 kb and 2.7 kb bands are detected. Genomic pattern of the ER eene in Drimarv
breast tumors.
A series of ER-negative
and ER-positive tumors were analyzed to detect any qualitative or quantitative changes in the ER gene. Some of the results obtained are illustrated in Fig. 2A, where a series of ER-positive
tumors were analyzed by Southern blots with the HE0
probe. In the same figure the results of hybridization
with the rRNA
and fibronectin
probes are shown (Figs. 2B and 2C respectively). When the ratio between the HE0 and rRNA bands for placenta was taken as the unity, the ratio for tumors #9, #14, #4, #7 and #6 were 3.0, 1.6, 1.0, 1.0 and 1.0, respectively. When the ratio between the HE0 and the fibronectin bands for placenta was taken as the unity, identical results were obtained. As a consequence, the first ratio was used for the analysis of every tumor. The relevant clinical characteristics of the patients studied, the ER gene amplification
and the biochemical
determinations
of the ER and PgR are indicated
in Table 1. The results obtained, shown in Table 1 and illustrated in Fig. 2, demonstrate that: 1) the ER genomic pattern was similar in both ER-negative and ER-positive
tumors;
2) a single copy of the ER-gene was found in every analyzed 603
Vol.
166, No. 2, 1990
BIOCHEMICAL
A kb 1 2 3 4 5
kbl
AND BIOPHYSICAL
B 2 34
RESEARCH COMMUNICATIONS
5 A
kb 1 2 3 4 5 6
6.5 * 7.6: 6.9 5.0 -
3.63.1 2.7 -
B 6.2, 4.5-
0
0
1
2
C 4.5 -D
Figure 1, ER geneanaivsisin human breastcancer ceil lines MCF-7 and MDA, placentaand normal endometrium.GenomicDNAs from thesehumantissueswere digestedwith EkoR I, electrophoresed,blotted and hybridized with plasmidsHE0 (A), HE14 (B), rRNA (C) and fibronectin (D). 1: placenta;2: MCF-7; 3: MDA; 4: endometrium#l; 5: endometrium#2. FiPure 2. ER Peneanalvsisin human breast cancer. DNA samplesfrom primary breast tumors were restricted with EcoR I, electrophoresed,blotted and after hybridizing with probe HEO (A), they were rehybridizedwith probe for rRNA (B) and fibronectin (C). Five representativesamplesof the tumorsanalyzedare shown. 1: #9; 2: #14; 3: #4; 4: #7; 5: #6; 6: placenta.
ER-negative tumor (8/8); 3) in some ER-positive tumors (6/14) an amplification of the ER-gene was detected. Even when present, the detected amplification was relatively low (up to 3- fold) and was not directly related to the ER content. DISCUSSION The discrimination of human breast cancers into ER-positive and ERnegative has great prognostic value. Thus, about 60% of patients with ER-positive tumors respond to hormonal treatments as compared to only 510% of patients with ER-negative tumors (21). It is therefore evident that the ability of human breast cancer to express ER is associatedwith a different biological behavior. The reasons underlying that increased expression are not clearly understood. A priori, several possibilities may be envisaged: a) an increased gene dosage of the ER gene; b) an increased transcriptional activity of that gene; c) modifications at the translational or post-translational level. In this paper we have provided evidence strongly suggesting that a higher gene dosage may play a role in the increased amount of ER found in some ER-positive tumors. This is based in the fact that in 6/14 ER-positive tumors a 604
Vol.
BIOCHEMICAL
166, No. 2, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
TableI Correlation betweenER content andER genecopynumber Patient
Age
Clinical Stage
:
5:
:I
445 397
241 0
1.6
ii 5
:t 57;
&I IIIb
308 275 200
207 60 71
kil:o 1.0
IIa I
141 187
1:s
::i
:
ER PgR (fmoles/mg prot)
ER (no of copies)
;
2
IIb
103
241 348
:::
11 10
2:
Ihb
2; 32
169 10
E+
12 13 14
5: 68 35
i;
14 12 13
0
IIIa II
208
E 1:6
15 16
61 ::
IIIb II
::
i
1.0 1.0
17 18 19
i;’
I IIIa
0 0
0 i
::“o 1.0
Z* 22
2: 83
rr :::
0i 0
0 ii
::i 1.0
The clinical characteristics of the patients studied and the ER and PgR values are indicated. After extraction and digestion with EcoR I, DNA was electrophoresed, blotted and the filters were hybridized sequentially with the HE0 and rRNA probes as described under Methods. For estimating the ER gene copy number, the 3.6 kb band of the ER-gene and the 6.2 kb band detected with the rRNA probe were densitometrically measured. The ratio ER/rRNA obtained for placenta was taken as 1. * Axillary metastasis.
certain degree of amplification of the ER gene is found, asopposed to a normal level in the 8 ER-negative tumors analyzed. Although the level of amplification detected was relatively low (1.6 to 3.0-fold), it should be taken into account that these are minimal values since there is a significative contribution of normal fibroblasts and lymphoid cells in human breast cancer tissues. However, the possibility that the increased dosage of the ER gene is due to a partial or complete duplication of chromosome 6 should also be considered. Although the number of casesanalyzed are too few to draw any definite conclusions, no correlation was found between the level of gene amplification and ER expression. The mechanism of ER expression in breast cancer appears to be quite complex. Thus, Ballare et al. (22) have recently found that in ER-positive tumors about 70% of the DNA-synthesizing cells are ERnegative, suggestingthat ER are not present in the most primitive breast cancer cells, but are acquired as the differentiation proceeds. The same authors have found that the stability of the ER differs among ER-positive tumors, pointing to possible differences either in the ER molecule itself or in the cellular processingmechanisms. 605
Vol. 166, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Recently, several species of ER mRNA have been found in human breast cancers, suggesting additional mechanisms of variation in the ER content (23). It is noteworthy that no changesin the EcoR I restriction pattern of the ER gene have been detected in any of the analyzed tumors. Other authors arrived to the same results with enzymes like BamH I, Hind III, Sst I and Pst I, although with Pvu II two polymorphic alleles in the hormone binding domain of the ER gene were found (24). These results support the idea that there are no extensive chromosomal rearrangements or deletions in the ER gene, even in ER-negative tumors, where the incidence of chromosomal changesis considerably larger (8-10). ACKNOWLEDGMENTS This work has been supported by Grants from the Consejo National de Investigaciones Cientificas y Tecnicas (CONICET), Argentina and the Comision de Investigaciones Cientificas y Tecnicas de la Provincia de Buenos Aires, Argentina. J. M. is a Career Investigator and M. N. a Fellow of the former Institution. We are grateful to Dr. C. Levy (Hospital Ramos Mejia, Buenos Aires, Argentina) and Dr. C. Garbovesky (Hospital Municipal de Oncologia, Buenos Aires, Argentina) for the provision of the tumors. REFERENCES 1. Anderson J. N. (1984) in Biological Regulation and Development, Goldenberg, R. F. and Yamamoto, K. R., eds. (Plenum, New York), Vol. 3B, 169-212. 2. Walter P., Green S., Greene G., Krust A., Bornet J-M, Jeltsch J-M, Staub A., Jensen E., Scrace G., Waterfield M. and Chambon P. (1985) Proc. Natl. Acad. Sci. (USA) 82,7889-7893. 3. Green S., Walter P., Kumar V., Krust A., Bornet J-M., Argos P. and Chambon P. (1986) Nature 320,134-139. 4. Petersen 0. W., Hoyer P. E. and van Deurs B. 1987) Cancer Res. 47,5748-5751. 5. Jensen E. V., Polley T. Z., Smith S., Block G. Ii ., Ferguson D. J. and DeSombre E. R. (1975) in Estrogen Receptors in Human Breast Cancer, McGuire W. L., Carbone P. P. and Vollmer E. P., eds. (Raven Press,New York) Chapter 5,37-56. 6. Nishikura K., Ar-Rushdi A., Erikson J., Watt R., Rovera G. and Croce C. M. (1983) Proc. Natl. Acad. Sci. (USA) 80,4822-4826. 7. Slamon D. J., Godolphin W., Jones L. A., Holt J. A., Wong S. G., Keith D. E., Levin W. J., Stuart S. G., Udove J., Urlich A. and Press M. F. (1989) Science 244, 707-712. 8. Dressler L. G., Seamer L. C., Owens M. A., Clark G. M. and McGuire W. L. (1988) Cancer 61,420-427. 9. Hedley D. W., Rugg C. A., Ng A. B. P. and Taylor I. W. (1984) Cancer Res. 44, 5395-5398. 10. Devilee P., Thierry R. F., Kievits T., Kolluri R., Hopman A. H. N., Willard H., Pearson P. L. and CornelisseC. J. (1988) Cancer Res. 48,5825-5830. 11. Soule H. D., Vazquez J., Long A., Albert S. and Brennan M. A. (1973) J. Natl. Cancer Inst. 51,1409-1416. 12. Cailleau R., Young R., Olive M., Reeves W. J. Jr. (1974) J. Natl. Cancer Inst. 53, 661-674. 13. Resnicoff M., Medrano E. E., Podhajhcer 0. L., Bravo A. I., Bover L. and Mordoh J. (1987) Proc. Natl. Acad. Sci. (USA) 84,7295-7299. 14. Davis R. W., Thomas M., Cameron J., St. John T. P., Scherer S. and Padgett R. A. (1980) Meth. Enzymol. 65,404-411. 15. Reed K. C. and Mann V. A. (1985) Nucleic Acid Res. 13,7207-7221. 16. Kumar V., Green S., Staub A. and Chambon P. (1986) Embo J. 5,2231-2236. 17. Bowman L. H., Rabin B. and SchlessingerD. (1981) Nucleic Acid Res. 9, 49514967. 606
Vol.
166,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
18. Gutman A. and Kornblihtt A. (1987) Proc. Natl. Acad. Sci. USA) 84,7179-7182. 19. McDonell M. W., Simon M. N. and Studier F. W. (1977) 5 . Mol. Biol. 110, 119-
146. 20. McGuire 989.
W. L. and De La Garza M. (1973) J. Clin. Endocrinol.
Metab. 37, 986-
21. Edwards D. P., Chamness G. C. and McGuire W. L. (1979) Biochem. Biophys. Acta 560.457-486. 22. Ball&e C., Bravo A. I., Laucella S., Sorin I., Cerdeiro R., Loza J., Sousa Martinez F., Guman N. and Mordoh J. (1989) Cancer 64,842-848. 23. Murphy L. C. and Dotzlaw H. (1989) Mol. Endocr. 3,687-693. 24. Hill S. M., Fuqua S. A. W., Chamness G. C., Greene G. L. and McGuire W. L. (1989) Cancer Res. 49,145148.
607
