Cathepsin D gene of human MCF7 cells contains estrogen-responsive sequences in its 5′ proximal flanking region

Vol.

174,

January

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

31, 1991

Pages

CATHEPSIN

D GENE

RESPONSIVE Vincent

OF HUMAN

SEQUENCES

CAVAILLES, Unite

Received

COMMUNICATIONS

December

IN ITS

Patrick

MCF7

5'

816-824

CONTAINS ESTROGENPROXIMAL FLANKING REGION CELLS

AUGEREAU,

and Henri

ROCHEFORT

Hormones et Cancer (U 148) INSERM Universite de Montpellier I 60, rue de Navacelles 34090 MONTPELLIER FRANCE 11,

1990

Cathepsin D is a lysosomal protease produced and secreted in excess by most human breast cancer cells. In MCF7 cells, estrogens stimulate cathepsin D expression at the mRNA level y& a mechanism independent of de m protein synthesis. We have isolated the human cathepsin D gene and its 5' flanking sequences from a MCF7 genomic library. To demonstrate its transcriptional estrogen regulation, we constructed chimeric recombinants bearing different fragments of the cathepsin D gene 5' proximal region inserted in front of the bacterial chloramphenicol acetyl transferase reporter gene. By transient cotransfection with the estrogen receptor expression vector into MCF7 cells, we showed that a 240 bp fragment located in the 5' proximal region of the gene was able to mediate transcriptional estrogen activation. This induction was concentration-dependent and suppressed by the antiestrogen ICI 164, 384. 0 1991AcademicPress, Inc. Estrogens stimulate growth of human metastatic breast cancer cell lines containing estrogen receptor (ER) (1) I following induction of several proteins (2) - Some of these proteins, such as growth factors and proteases, have been proposed to play a decisive role in promoting tumor growth and metastasis

(3) .

High levels of an acidic lysosomal proteinase, cathepsin associated with an increased risk of developing D, are metastases in patients (4) and with increased metastatic ability in experimental systems, following stable transfection of its cDNA into tumoral cells (5). The determination of overexpression and mechanisms involved in cathepsin D gene hormonal regulation may therefore be crucial in understanding steps of hormonal carcinogenesis. Previous studies have shown that in ER-positive human cathepsin D mRNA is specifically breast cancer cell lines,

Vol.

174,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

induced by estrogens via a mechanism which is independent of de novo protein synthesis (6). Moreover, run-on nuclear assays and the use of transcription inhibitors suggest a direct stimulation of transcription. However, cathepsin D regulation appears to be particularly complex. Its mRNA is also induced indirectly by different growth factors in MCF7 cells (7) and regulation by steroid hormones varies according to the tissue, since in normal endometrium cathepsin D is induced by progesterone but not by estrogens (8,9). Moreover, this gene exhibits a different sensitivity to antiestrogens in antiestrogen-sensitive and -resistant human breast cancer cell lines (10). Activation of transcription by steroid hormones is mediated by specific nuclear receptors acting as ligand activated transcription factors interacting specifically with cognate responsive elements (HRE) generally located in the 5' flanking region of the gene (11). Among the estrogen-induced responses in mammalian cells, only a few have been characterized as being due to a direct stimulation of gene expression and the structures of the estrogen responsive according to the gene. elements vary In order to specify whether the cathepsin D gene also contains regulatory sequences responding to estrogen, we cloned and characterized the human gene isolated from MCF7 breast cancer cells and studied the estrogen-dependent expression of recombinants containing upstream regions of the cathepsin D gene. MATERIALS

AND METHODS

Isolation of aenomic DNA clones Two ?, phage genomic DNA libraries, established from MCF7 cell DNA digested by Eco RI or Hind III were used to clone the cathepsin D gene. Approximately 8.105 plagues were @we screened using the entire cathepsin D cDNA as a probe (12). Restriction endonuclease mapping Phage DNAs were isolated from the positive plagues and restriction endonuclease analysis was carried out. We first used the "Lambda Mapping SystemI' from Amersham according to the manufacturer's protocol. In parallel, we also performed analysis of the totally digested DNA by Southern Blot hybridization either with total cDNA or with more 5' specific probes i.e. the 177 bp S'/Sca I fragment or the 285 bp Barn HI fragment (Fig. 1) Plasmid

the

recombinant9

Different regions of the HSV thymidine kinase (tk)

cathepsin promoter 817

D gene were fused in the pBL-CAT8+vector

to

Vol.

174,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

The 7 kb Hl-Bl fragment (Fig. 2A) isolated as a Barn HI (13). fragment from the HO82 clone was introduced in the Barn HI site of the pBL CAT8+ vector (clone CDl) ; the 5.2 kb Xbl-Xb2 fragment (clone CD31) and the 4kb Hl-Xbl fragment isolated as a Xba I fragment (clone CD36) were subcloned into the Xba I site of the pBL CAT8+ ; the 2.3 kb X4-X5 fragment (clone CD46) and the 0.24 kb X2-X3 fragment (clone CD24) were introduced into the same Xba I site by blunt-end ligation after filling in with T4 DNA polymerase. As a positive control for estrogen induction, we used the Vit-tk-CAT recombinant which contains the -331/-87 fragment of the Xenopus vitellogenin A2 gene inserted in front of the tk promoter (13) and the pS2 (-llOO/+lO) CAT recombinant (14). The parental plasmid pBL-CAT8+ served as a negative control in transfection experiments. The HE0 expression plasmid containing the full-length ER cDNA (15) was generally cotransfected. DNA transfection and transient exoression assay MCF7 cells were maintained in Dulbecco's Modified Eaule's medium (DMEM) supplemented with 10% fetal calf serum (Gibco) and switched to medium containing charcoal-treated serum (FCSDCC) without phenol red 1 week before transfection. were the Transfections performed by calcium-phosphate precipitation method (16). DNA precipitates (500 ng of CAT recombinant and 500 ng of HE0 for a total of 5 c(g DNA) were added to monolayers of subconfluent cells plated 24 h earlier in six multiwell plates. Cells were then incubated at 37°C in 5% CO2 for 24 h. Subsequently, cells were washed and half were treated with 1OnM estradiol for 24 h. Cell extracts were prepared by 5 cycles of freeze-thawing followed by centrifugation at 10 000 x g for 15 min. Protein concentrations were determined by the Bradford assay (BioRad) and CAT activities were quantified as previously described (17). RESULTS Isolation

and marina

To

clone

libraries from

were of

previously (11-16

kb)

containing method.

human

D oene

cathepsin

D

in

the

by

Eco RI

MCF7

total to

of

DNA,

generate with

800

000

and

gene,

X dash

these

or

D cDNA

plagues

the

Hind

genomic

phage Hind

molecular were

D gene by the giving a positive

3 for

two

III

DNA

(Stratagene),

III.

By Southern

restriction

high

cathepsin

the cathepsin Seventeen clones

RI library isolated.

cathemin

established

detected

total

human

DNA digested

shown

A

Eco and

the

MCF7 cell

analysis

of

AND DISCUSSION

enzymes weight

were

fragments

(12). screened

for

plague signal

library)

clones

hybridization (14 for

were

the

rescreened

Clones from the two libraries overlapped and spanned about 20 kb. As indicated by hybridization with the 5' specific probes (5l/sca I and Barn HI fragments ; Fig. l), clones from the

Hind

III

library

extended

further 818

upstream,

compared

to

Vol.

174,

No.

2, 1991

BIOCHEMICAL

X4 Hl

El

5' I

I

AND

BIOPHYSICAL

01 SC1

812 B3Kl

RESEARCH

COMMUNICATIONS

85

KG?

X6

B‘lE2

Ii2

I I

I 3'

HO82 genomic clone -

500 bp

X4

-

J ’ ”

5'

cDNA

SC1 Kl SC2 83 K2

81

85 X6

Ii

‘-

I

3'

100bp 5’ specific probes

Picrure Restriction

1 map

of

the

HO82

human

D genomic

cathepsin

clone

Restriction endonuclease map of the HO82 clone was carried out as described in the Materials and Methods. The enzymes used were Hind III (H), Eco RI (E), Xma III Sites are numbered (Xl, Barn HI (B), Sea I (SC) and Kpn I (K). according to their position in the cathepsin D gene and those which are present in the cDNA are shown in bold characters. and the +1 indicates the initiation of translation translated sequences of the cDNA are shown by the shaded box. The position of the two 5' probes (the 177 bp 5'/Scl and the 285 bp Bl/B3 fragments) is also indicated.

clones from the Eco restriction endonuclease

RI

library. analysis

We therefore of one of these

Comparison clones (clone H082). cathepsin D cDNA, and hybridization allowed

us to

find

coding

region

of

that

the

the

cDNA

sequences (Fig. 1). By Southern analysis hybridization

pattern

for the flanking with MCF7 total

absence

of major

Localization

of

activation

of

To test located

the

secruence

map with

with

and

about the

with

fragments DNA (not

rearrangements

this 5'

HO82 clone

using

obtained

(except obtained

of

most

4

5'

kb

of

mediatina

of

probes,

contained

of

the

upstream

cDNA probe,

the

HO82 DNA was identical

of the shown),

during

that

specific

complete the

performed Hind III

HO82 clone) to thus indicating

the

cloning the

that the

procedures.

estrosen

induced

transcriDtion

the

upstream

possible

responsiveness

of

sequences

cathepsin D gene, we constructed Various fragments of the HO82 recombinants.

several chimeric clone were inserted Tk promoter and the

of

estrogen

the

in the pBL CAT8i vector bacterial CAT gene (Fig. 819

in front 2A).

of the

HSV

Vol.

174,

No.

BIOCHEMICAL

2, 1991

x4

A

I

I

BIOPHYSICAL

x5

82 I

II

RESEARCH

COMMUNICATIONS

8cz

61 SC1

x2 x1x3

ml

Hl 5'

AND

Ial2

I

I

3’

HO62 genomic

clone

B’region Y 500 bp

+1 CD1 CD31 CD 35 CD46

__-- 3’ Tk promoter B

+HEO CD1

CD36

CD46

CD24

PBL

PS2

2

Localization

B.

CD31

-+-+-+-+++++-+++

Fiuure

A.

CAT gene

-HE0

E2

pBLCATB+

of

sequences

mediating

estrogen

reSPOnSiVeneSS

representation of the reporter recombinants. Schematic clone 5' region were Various fragments of the HO82 inserted in the pBLCAT8+ vector in front of the HSV Tk chloramphenicol acetyl promoter and the bacterial transferase (CAT) gene. performed after showing the CAT assay Autoradiogram transfection of MCF7 cells with the clones CDl, CD31, CD36, CD46 or CD24 as described in the Materials and receptor expression vector Methods. The human estrogen (HEO) was cotransfected (+) except for lanes 1 and 2. DNA precipitates contained 500 ng of reporter plasmid and 500 ng of HE0 vector. Cultures were treated for 24 h with 10 nM estradiol (E ) or with 0.1% ethanol alone (C) and aliguots of ce!?l extracts (100 c(g proteins) were assayed for CAT activity. Transfection controls were performed with the recombinants ; pBL) and the pS2 CAT plasmid pBLCAT8+ (negative control inducible by estrogens (pS2). 820

Vol.

174,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

We first transfected in MCF7 human breast cancer cells, the CD1 clone which contained the entire cloned 5' region, the first exon and the first intron in the 7 kb insert. As shown in Fig. 2B (lanes l-2), no induction of CAT activity was observed, whereas the endogeneous cathepsin D gene was stimulated as indicated by immunoassay of the secreted cathepsin D (not shown). Cotransfection with the human estrogen receptor expression vector HE0 resulted in the appearance of a 4-Fold transcriptional stimulation (lanes 3-4), whereas the parental pBL CAT8+ plasmid was not affected by the hormonal treatment (lanes 13-14). This result indicates that the 7 kb insert of the CD1 clone contained estrogen responsive sequences (ERE). It also suggests that the concentration of endogeneous ER may not be sufficient, relative to the amount of cathepsin D reporter DNA introduced into MCF7 cells. The position of the estrogen responsive element(s) was then specified by transfecting into MCF7 cells other reporter recombinants containing smaller cathepsin D gene fragments. As shown in Fig. 2B (lanes 5-12), estrogen induction was noted for the recombinant CD31, whereas recombinants CD36 and CD46 were unresponsive in the same conditions. This suggested that the regulatory sequences were located between the Xma III sites Xl and X4. In fact, transfection of the CD24 clone localized estrogen responsive sequences in a 240 bp fragment located at -123/-364, when taking +l as the Adenine of the first ATG used for translation. A similar induction of CAT activity was also obtained in the same conditions after cotransfection of the CD24 and HE0 vectors in the ER-negative HeLa cell line (not shown). Specificity of estroaen induction Transcriptional regulation was mediated by the ER as indicated by dose-reponse experiments performed with the CD24 induction was obtained with clone (Fig. 3A). Half maximal approximately 1nM estradiol and maximal effect with 1OnM. The higher concentrations, compared to those required to regulate the endogeneous cathepsin D gene (maximal induction for O.lnM), is explained by the fact that the ER encoded by the HE0 plasmid has a loo-fold lower affinity for its ligand than wild type ER due to a mutated val in the estrogen binding domain (18). Finally, we analyzed the effects of the antiestrogen ICI 164,384 on the estrogen regulation of clone CD24 (Fig. 3 B). 821

Vol.

174,

No.

0

2, 1991

-12

-11

-10

Log estradiol

BIOCHEMICAL

-9

-6

Fiaure 3 Dose-dependence A.

B.

-7

concentration

(lanes

RESEARCH

B

pBLCATB+

Vit-tk-CAT

E2 ICI

-+--+--+-+ - +-

COMMUNICATIONS

CD24

-

+-

5

6

-

+

8

910

+

-6

2

1

(M )

ICI

3

4

response 164,384

7

and effects

of

MCF7 cells were cotransfected with the HE0 and CD24 clones as described in Fig. 2. with Cells were stimulated 24 h after transfection increasing concentrations of estradiol and CAT activity was determined on aliquots of cell extracts as described in the Materials and Methods and the autoradiogram was scanned for quantification. (18) MCF7 cells were cotransfected with the HEGO vector and either control recombinants (pBL CAT8+ or vit-tk CAT) or the CD 24 clone. After one day of culture, cells were stimulated with 1 nM estradiol (E2), 500 nM ICI 164,384 alone (C). (ICI) I 1 nM E2 + 500 nM ICI or with ethanol CAT assays were performed as described in Fig. 2.

completely was

BIOPHYSICAL

of the transcriptional the antiestrogen

This compound behaved stimulate transcription It

AND

suppressed also

as a pure antagonist since it did not when added alone (lanes 7 and 9) and it estradiol

ineffective

stimulation

on the

transcription

(lanes of

8 and

control

10).

vectors

3 and 6).

To conclude, we cloned MCF7 cells and demonstrated by estrogens -123/-364 progress that of sequences estrogen from the (20) and progesterone

of

is the

mediated initiation

the human cathepsin that its transcriptional by a 240 bp fragment, codon. Further

D gene from regulation located at studies

are

in

to determine the structure of the ERE(s) compared to other estrogen-regulated genes, and the nature of which may cooperate in this activity. In mammalian responsive cells, the structure of EREs may diverge consensus sequence (19) by one base, as in the the oxytocin genes (21) and by more, as in receptor (22) and creatine kinase genes (23). 822

pS2 the

Vol.

174,

No.

BIOCHEMICAL

2, 1991

Molecular cathepsin physiological

mechanisms

D gene

AND

the

of

expression

and pathological

BIOPHYSICAL

can

now

RESEARCH

hormonal be

studied

COMMUNICATIONS

regulation in

of

different

conditions.

Acknowledsments This research was supported by the "Institut National de la Recherche Medicale", the "Centre National de la Recherche Scientifique", the "Faculte de Medecine de Montpellier" and by a fellowship granted to V. Cavailles by the "Fondation pour la Recherche Medicale". We are grateful to Prof. Chambon for the gift of pBL pS2-CAT and Vit-tk-CAT recombinants and to F. Janvier CAT8+, and C. Gros for typing the manuscript. We thank C. Gaudelet for for performing excellent technical assistance and S. Roudani some of the transfection experiments.

REFERENCES 1.

2. 3. 4.

5. 6. 7. 8. 9. 10. 11. 12.

13. 14. 15.

16. 17.

Lippman, M-E., Bolan, G. and Huff, K. (1976) Cancer Res. 36, 4595-4601. Chalbos, D., Vignon, F., Keydar, I. and Rochefort, H. (1982) J. Clin. Endocrinol. Metab. 55, 276-283 Rochefort, H., Capony, F., Garcia, M., Cavailles, V., Freiss, G., Chambon, M., Morisset, M., Vignon, F. (1987) J. cell. Biochem. 35, 17-29. Spyratos, T., Maudelonde, T., Brouillet, J.P., Brunet M., Defrenne, A., Andrieu, C., Hacene, K., Desplaces, A., Rouesse, J., Rochefort, H. (1989) The Lancet ii, 8672, 1115-1118. Garcia, M., Derocq, D., Pujol, P., Rochefort, H. (1990) Oncogene, in press. Cavailles, V., Augereau, P., Garcia, M., Rochefort, H. (1988) Nucl. Acids. Res. 16, 1903-1919. Cavailles, V., Garcia, M., Rochefort, H. (1989) Mol. Endocrinol. 3, 552-558. Elangovan, S., and Moulton, B.C. (1980) J. Biol. Chem. 255, 7474-7479. Touitou, I., Cavailles, V., Garcia, M., Defrenne, A., Rochefort, H. (1989) Mol. Cell. Endocrinol. 66, 231-238. Westley, B., F.E.B., Brown, A.M.C., Krust, A., MaYI Chambon, P., Lippmann, M-E., Rochefort, H. (1984) J. Biol. Chem. 259, 10030-10035. Beato, M. (1989) Cell 56, 335-344. Augereau, P., Garcia, M., Mattei, M-G., Cavaillbs, v., Depadova, F., Derocq, D., Capony, F., Ferrara, P., Rochefort, H. (1988) Mol. Endocrinol. 2, 186-192. Klein-Hitpass, L., Schorpp, M., Wagner, U., and Ryffel, G.V. (1986) Cell, 46, 1053-1061. Kumar, V., Green, S., Stack, G., Berry, M., Jin, J-R., and' Chambon, P. (1987) Cell, 51, 941-951. Walter, P., Kumar, V., Krust, A., Green, S., Bornert, J-M., Argos, P., and Chambon, P. (1986) Nature 320, 134139. Graham, F-L., and Van der Eb, E.J. (1973) Virology 52, 456-467. Gorman, C-M., Moffat, L.F. and Howard, B.H. (1982) Mol. Cell. Biol. 2, 1044-1051. 823

Vol.

18. 19. 20. 21. 22. 23.

174,

No.

2, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Tora, L., Mullick, A., Metzger, D., Ponklikitmongkol, M., Park, I., and Chambon, P. (1989) Embo J. 8, 1981-1986. Walker, P., Germond, J.E., Brown-Luedi, M., Girel, F., and Wahli, W. (1984) Nucleic Acids Res. 12, 8611-8626. Berry, M., Nunez, A.M., and Chambon P. (1989) Proc. Natl. Acad. Sci USA, 86, 1218-1222. Richard, S., and Zingg, H.H. (1990) J. Biol. Chem. 265, 6098-6103. Kastner, P., Krust, A., Turcotte, B., Stropp, U., Tora, L Gronemeyer, H. and Chambon, P. (1990) Embo. J. 9, 1663-1614. Pentecost, B.T., Mattheiss, L., Dicker-man, H.W., and Kumar S.A. (1990) Mol. Endocrinol. 4, 1000-1010.

824