566
BIOCHEMICAL SOCIETY TRANSACTIONS
5. Thorpe, S. M., Lykkesfeldt, A. E., Vinterby, A. & Lonsdorfer, M. ( 1 986) C'uncer Rex 46,425 Is-4255s 6. Wittliff, J. L. ( 1 979) Mol. Aspects Med. 2, 395-437 7. Wrange, O., Humla, S., Ramberg, J., Nordenskjold, B. & Gustafsson, J. A. ( 1 980) Rec. Res. Cunc. Hes. 91,32-40 8. Wiehle, R. D. & Wittliff, J. L. (1984) J. Chromurogr. 297, 3 13-321
9. Wrange, 0..Norenskjold. B. & Gustafsson, J. A. ( 1978) And. Biochem. 85,46 1-475 10. Moncharmont, B., Anderson, W. L.. Rosenbcrg, 13. & Parikh, I. ( 1 984) Biwhemisiry 23,3907-39 I 2
Received 24 November I 989
Oestrogen and anti-oestrogen induction of specific gene expression in human breast cancer cells CHRISTINA M. W. CHAN and CHRIS D. GREEN Department of Biochemistry, University of Liverpool, I? 0. Box 147, Liverpool L693BX, U.K. Approximately one-third of all human breast tumours is hormone-responsive, i.e. the growth of the tumour cells is stimulated by oestradiol [ 11. The metastases of such tumours are commonly treated with anti-oestrogenic drugs, e.g. Tamoxifen 121. It is well established that anti-oestrogens compete with oestradiol for binding to the oestrogen receptor [3] and that their potency is related to their affinity for the receptor. However, only approximately 55% of breast tumours containing oestrogen receptors are growth-inhibited by anti-oestrogens. Furthermore, it has been shown that, for certain responses, anti-oestrogens may act as full or partial oestrogen agonists. The growth of the human breast cancer cell lines, MCF-7 and ZR-75-1, is oestrogen dependent in culture [4, 51 and may serve as a model of human hormone-responsive breast cancer in vitro. We possess two cDNAs, pS2 [6]and pLIV-1 (71 that were isolated on the basis of their induction by oestradiol in MCF-7 and ZR-75-1 cells, respectively. We have examined the responses of both genes to antioestrogens and to other steroid hormones in both MCF-7 and ZR-75-1 cells. Cells were grown for 6 days (three medium changes) in Phenol Red-free, RPMl 1640 medium with 10% (v/v) dextran-charcoal-stripped newborn calf serum, to deplete them of steroid hormones. The cells were then transferred to the same medium supplemented with either oestradiol ( l o - ' M) or 5a-dihydrotestosterone ( l o - '
M ) or dexamethasone ( l w 7 M) or tamoxifen ( l o - " ) o r 4-hydroxytamoxifen ( l o - ' M ) or t o medium without addition. After culture for 3 further days, the cells were harvested and their total RNA isolated. The RNA was transferred to a nylon membrane (Hybond N, Amersham) by the 'dot blot' procedure and hybridized to "P-labelled pLIV- 1 cDNA. After hybridization (45"C, 18 h), the membrane was washcd to high stringency (final wash, 0 . 1 x SSPE, 20°C 171)and then autoradiographed (Fig. I). The same membrane was then stripped of probe and rehybridized to "P-labelled pS2 cDNA (Fig. 1 ). It can be seen that both pLIV-I and pS2 are expressed in an oestrogen-responsive fashion in both cell-lines (Fig. 1 a and b, tracks 1-4, 6, 9). Expression o f both genes is also induced in MCF-7 cells by 5 a-dihydrotestosterone (Fig. 1 a and 6, track 5 ) . Interestingly, pLIV-I is strongly induced by the glucocorticoid, dexamethasone, but pS2 is unresponsive to this hormone (Fig. l a and b, track 7). The two genes also differ in their response to anti-oestrogens. pLIV-I is markedly induced by tamoxifen (Fig. l a , track X), whereas the level of pS2 expression in tamoxifen-treated ZR-75- 1 cells (Fig. 1b, track 8) is indistinguishable from that in control cells (Fig. 1b, track 1). 4-Hydroxytamoxifen stimulates a slight increase in pLIV-1 (Fig. l a , track l o ) , but is without detectable effect on pS2. Using a more sensitive hybridization assay, May & Westley [S]were able to detect a weakly oestrogenic effect of tamoxifen and 4-hydroxytarnoxifen upon pS2 expression in MCF-7 cells. pLIV- 1 and pS2 therefore represent two genes expressed in the same cells and similarly responsive t o the hormones
-
.. :>' .
(b)
?
2
3
4
0 .
0
0 .
m .
I.
0 .
o r
0
*
.
--3
0
1
d o 0 9
5
6
7
8
9 1 0
1
2
3
4
5
6
7
8 9 1 0
Fig. 1 . RNA dot-blot atiulysis of ZR-75-1 arid MCF7cells treated with hormones und anti-oestrogetis ( a )pLlV-1 probe; ( h ) pS2 probe: 1, ZR-75-1 cells, 6 days steroid depleted; 2, ZR75-1 cells, 6 days oestradiol ( l o - ' M); 3, MCF-7 cells, 6 days steroid depleted; 4, MCF-7 cells, 6 days oestradiol ( 1 W XM); 5, MCF-7 cells, 6 days steroid depleted, 3 days dihydrotestosterone ( l o - ' M); 6, MCF-7 cells, 6 days steroid depleted, 3 days oestradiol ( l o - ' M); 7,ZR-75-1 cells, 6 days steroid depleted, 3 days dexamethasone ( 10F7M);8,ZR-75- 1 cells, 6 days steroid depleted, 3 days tamoxifen ( l o - " M); Y, ZR75-1 cells, 6 days steroid depleted, 3 days oestradiol ( lo-' M); 10, ZR-75- I cells, 6 days steroid depleted, 3 days hydroxytamoxifen ( l o - ' M).
633rd MEETING, LONDON
567
oestradiol and dihydrotestosterone. The two genes, however, show profound differences in their responses to other hormonal and anti-hormonal agents. The responsiveness of pLIV- 1 to dexamethasone presumably indicates the presence of a glucocorticoid response element 191 in association with this gene, whereas the pS2 gene lacks such a regulatory element. The marked difference in sensitivity t o anti-oestrogens cannot be explained in this manner. There is evidence for physical and functional differences between oestrogen/ receptor complexes and anti-oestrogen/receptor complexes [ 10. 1 1 I. If different genes possess different abilities to distinguish between the two types of receptor complex, either at the level of receptor/DNA binding or of events subsequent to binding, this may be the basis for their individual responses. We arc grateful t o Dr P. Chambon for supplying the pS2 cDNA. This work was supported by a grant from the North West Cancer Research Fund.
I . Henderson, 1. C. & Canellos, G. P. (1980) New E r g / . J . Med. 302. 7X-90
2. Jordan, V. C.. Fritz. N. F.; & Gottardis. M. M. ( 10x7) J. S t c w i c l Hiochem. 21.493-498 3. Jordan, V. C. ( 1986) E.strogm/Atriic..s/ro~i~tr Ac,iiorr urrd Ilriwst C'uncer Therupy. University o f Wisconsin Press. Madison 4. Berthois. Y..Katzenellenbogen, J. A. & Katzenellenbogen. B. S. (1986) /'roc. Null. Acud. Sci. U..S.A.83, 2496-2500 5. Darbre, P., Yates, J., Curtis, S . & King. K. J. B. ( I 983) C i r r m r Hex 43,349-354 6. Masiakowski. P.. Breathnach, R.. Bloch. J.. Gannon. F.. Krust. A. & Chambon. P. ( 19x2) Niich4c. Acids He.\.10. 7x95-7903 7. Manning, D. L.. Daly. R. J.. Lord, P. G., Kelly. K. E & Grccn. C. D. (1988) Mol. Ci.llEntlocritio1. 59. 205-2 12 8. May, F. E. B. & Westley, B. R. (1987) J. H i o l . C'lieni. 262, 15894-15899 9. Beato. M. ( 1 9x9) C'c.11 (C'umbridgc, Muss.) 56. 335-344 10. Geier, A,, Beery, R.. Haimsohn. M. & Lunenfeld, B. ( 1987) J. Steroid Hiocliem. 28.47 1-478 I 1. Martin, P. M.. Berthois, Y. & Jensen, E. V. ( I98X) /'roc. NutI. A c ~ dS. C ~ .U.S.A. 85,2533-2537
Received 23 November 19x9
Effect of changes in 5' coding sequence on level of expression of ovine growth hormone cDNA in Escherichia coli attention. Poor bacterial expression o f such hormones has been markedly improved by manipulation o f the sequencc 5' or 3' to the initiation codon [e.g. 1-51. It has been suggested that the original poor expression in thesc systems is caused by the formation of secondary structures in the mRNA which inhibit the initiation of translation. In most cases the nature of such secondary structures remains unclear. To investigate this problem we have previously constructed a series of plasmids in which thc cDNA sequence
A. JAMIL SAMI, 0.CARY L WALLIS and M IC HA EL WA LLI S Biochemistry Luhorutory, School of Biological Sciences, Utiivemity of Sussex. Fulmer, Rrightori BNI SQG, U.K . Production of biologically active growth hormones (GH) by recombinant DNA technology is receiving considerable Abhreviations used: GH, growth hormone; o. ovine. Plasrnid
v
... ...
pOGHel04
ATGACCATCCCAGCCATG
pOGHelO5
ATCACTCCCAGCCATG
oOGHelO6
ATGACCAACTTCCCAGCCATG
pOGHelO7 pOGHelO8
oGH variant content (as % total cell protein)
Nucleotide seauence
v
15
v .... v ATGACCGACTTCCCAGCCATG . . . v ATGACGCACTTCCCAGCCATG
22
...
3
v
pOGHelO9
ATGACCATTCCCGGGGACTTCCCAGCCATG . . . .
10
pOGHel10
ATGACAATTCCCGGGC;ACTCCCAGCCATG
....
3
pOGHel11
ATGACCATGATTACAGCCATG .
pOGHell2
ATCACCA~ATTCCCGGCGACTTCCCAGCCATG
pOGHell3
ATGACCATGA~'AATTCCCGGGGACMCCCCAGCCATG. . . .
pOGHelO3
ATGACCATGATTACCAAT~AGCCATG.
pOGHell4
ATGACCAI'GATTACGAATTCGGACTTCCCAGCCATG
pOGHel01
v
v
..
11
v
,
...
..
v ... v ATGACCATGAI'TACGAATTCCCGCGACTCCCAGCCATG . . .
11 11 15 12 27
Fig. 1. Effect of nircleotide seqirerice in expression plasmids on level of expression of oGH vuriunts The sequence of bases 3' to the initiation codon and the oGH content of induced bacteria containing the corresponding plasmid were determined as described in the text. Bases in italics are derived from pUC8 and the remainder are from the oGH coding sequence; V shows the site of deletions; underlined bases show differences from the B-galactosidase sequence in the first eight bases. Data for pOGHe 10 1 are from Wallis & Wallis 16, 8). Vol. I8
BIOCHEMICAL SOCIETY TRANSACTIONS
5. Thorpe, S. M., Lykkesfeldt, A. E., Vinterby, A. & Lonsdorfer, M. ( 1 986) C'uncer Rex 46,425 Is-4255s 6. Wittliff, J. L. ( 1 979) Mol. Aspects Med. 2, 395-437 7. Wrange, O., Humla, S., Ramberg, J., Nordenskjold, B. & Gustafsson, J. A. ( 1 980) Rec. Res. Cunc. Hes. 91,32-40 8. Wiehle, R. D. & Wittliff, J. L. (1984) J. Chromurogr. 297, 3 13-321
9. Wrange, 0..Norenskjold. B. & Gustafsson, J. A. ( 1978) And. Biochem. 85,46 1-475 10. Moncharmont, B., Anderson, W. L.. Rosenbcrg, 13. & Parikh, I. ( 1 984) Biwhemisiry 23,3907-39 I 2
Received 24 November I 989
Oestrogen and anti-oestrogen induction of specific gene expression in human breast cancer cells CHRISTINA M. W. CHAN and CHRIS D. GREEN Department of Biochemistry, University of Liverpool, I? 0. Box 147, Liverpool L693BX, U.K. Approximately one-third of all human breast tumours is hormone-responsive, i.e. the growth of the tumour cells is stimulated by oestradiol [ 11. The metastases of such tumours are commonly treated with anti-oestrogenic drugs, e.g. Tamoxifen 121. It is well established that anti-oestrogens compete with oestradiol for binding to the oestrogen receptor [3] and that their potency is related to their affinity for the receptor. However, only approximately 55% of breast tumours containing oestrogen receptors are growth-inhibited by anti-oestrogens. Furthermore, it has been shown that, for certain responses, anti-oestrogens may act as full or partial oestrogen agonists. The growth of the human breast cancer cell lines, MCF-7 and ZR-75-1, is oestrogen dependent in culture [4, 51 and may serve as a model of human hormone-responsive breast cancer in vitro. We possess two cDNAs, pS2 [6]and pLIV-1 (71 that were isolated on the basis of their induction by oestradiol in MCF-7 and ZR-75-1 cells, respectively. We have examined the responses of both genes to antioestrogens and to other steroid hormones in both MCF-7 and ZR-75-1 cells. Cells were grown for 6 days (three medium changes) in Phenol Red-free, RPMl 1640 medium with 10% (v/v) dextran-charcoal-stripped newborn calf serum, to deplete them of steroid hormones. The cells were then transferred to the same medium supplemented with either oestradiol ( l o - ' M) or 5a-dihydrotestosterone ( l o - '
M ) or dexamethasone ( l w 7 M) or tamoxifen ( l o - " ) o r 4-hydroxytamoxifen ( l o - ' M ) or t o medium without addition. After culture for 3 further days, the cells were harvested and their total RNA isolated. The RNA was transferred to a nylon membrane (Hybond N, Amersham) by the 'dot blot' procedure and hybridized to "P-labelled pLIV- 1 cDNA. After hybridization (45"C, 18 h), the membrane was washcd to high stringency (final wash, 0 . 1 x SSPE, 20°C 171)and then autoradiographed (Fig. I). The same membrane was then stripped of probe and rehybridized to "P-labelled pS2 cDNA (Fig. 1 ). It can be seen that both pLIV-I and pS2 are expressed in an oestrogen-responsive fashion in both cell-lines (Fig. 1 a and b, tracks 1-4, 6, 9). Expression o f both genes is also induced in MCF-7 cells by 5 a-dihydrotestosterone (Fig. 1 a and 6, track 5 ) . Interestingly, pLIV-I is strongly induced by the glucocorticoid, dexamethasone, but pS2 is unresponsive to this hormone (Fig. l a and b, track 7). The two genes also differ in their response to anti-oestrogens. pLIV-I is markedly induced by tamoxifen (Fig. l a , track X), whereas the level of pS2 expression in tamoxifen-treated ZR-75- 1 cells (Fig. 1b, track 8) is indistinguishable from that in control cells (Fig. 1b, track 1). 4-Hydroxytamoxifen stimulates a slight increase in pLIV-1 (Fig. l a , track l o ) , but is without detectable effect on pS2. Using a more sensitive hybridization assay, May & Westley [S]were able to detect a weakly oestrogenic effect of tamoxifen and 4-hydroxytarnoxifen upon pS2 expression in MCF-7 cells. pLIV- 1 and pS2 therefore represent two genes expressed in the same cells and similarly responsive t o the hormones
-
.. :>' .
(b)
?
2
3
4
0 .
0
0 .
m .
I.
0 .
o r
0
*
.
--3
0
1
d o 0 9
5
6
7
8
9 1 0
1
2
3
4
5
6
7
8 9 1 0
Fig. 1 . RNA dot-blot atiulysis of ZR-75-1 arid MCF7cells treated with hormones und anti-oestrogetis ( a )pLlV-1 probe; ( h ) pS2 probe: 1, ZR-75-1 cells, 6 days steroid depleted; 2, ZR75-1 cells, 6 days oestradiol ( l o - ' M); 3, MCF-7 cells, 6 days steroid depleted; 4, MCF-7 cells, 6 days oestradiol ( 1 W XM); 5, MCF-7 cells, 6 days steroid depleted, 3 days dihydrotestosterone ( l o - ' M); 6, MCF-7 cells, 6 days steroid depleted, 3 days oestradiol ( l o - ' M); 7,ZR-75-1 cells, 6 days steroid depleted, 3 days dexamethasone ( 10F7M);8,ZR-75- 1 cells, 6 days steroid depleted, 3 days tamoxifen ( l o - " M); Y, ZR75-1 cells, 6 days steroid depleted, 3 days oestradiol ( lo-' M); 10, ZR-75- I cells, 6 days steroid depleted, 3 days hydroxytamoxifen ( l o - ' M).
633rd MEETING, LONDON
567
oestradiol and dihydrotestosterone. The two genes, however, show profound differences in their responses to other hormonal and anti-hormonal agents. The responsiveness of pLIV- 1 to dexamethasone presumably indicates the presence of a glucocorticoid response element 191 in association with this gene, whereas the pS2 gene lacks such a regulatory element. The marked difference in sensitivity t o anti-oestrogens cannot be explained in this manner. There is evidence for physical and functional differences between oestrogen/ receptor complexes and anti-oestrogen/receptor complexes [ 10. 1 1 I. If different genes possess different abilities to distinguish between the two types of receptor complex, either at the level of receptor/DNA binding or of events subsequent to binding, this may be the basis for their individual responses. We arc grateful t o Dr P. Chambon for supplying the pS2 cDNA. This work was supported by a grant from the North West Cancer Research Fund.
I . Henderson, 1. C. & Canellos, G. P. (1980) New E r g / . J . Med. 302. 7X-90
2. Jordan, V. C.. Fritz. N. F.; & Gottardis. M. M. ( 10x7) J. S t c w i c l Hiochem. 21.493-498 3. Jordan, V. C. ( 1986) E.strogm/Atriic..s/ro~i~tr Ac,iiorr urrd Ilriwst C'uncer Therupy. University o f Wisconsin Press. Madison 4. Berthois. Y..Katzenellenbogen, J. A. & Katzenellenbogen. B. S. (1986) /'roc. Null. Acud. Sci. U..S.A.83, 2496-2500 5. Darbre, P., Yates, J., Curtis, S . & King. K. J. B. ( I 983) C i r r m r Hex 43,349-354 6. Masiakowski. P.. Breathnach, R.. Bloch. J.. Gannon. F.. Krust. A. & Chambon. P. ( 19x2) Niich4c. Acids He.\.10. 7x95-7903 7. Manning, D. L.. Daly. R. J.. Lord, P. G., Kelly. K. E & Grccn. C. D. (1988) Mol. Ci.llEntlocritio1. 59. 205-2 12 8. May, F. E. B. & Westley, B. R. (1987) J. H i o l . C'lieni. 262, 15894-15899 9. Beato. M. ( 1 9x9) C'c.11 (C'umbridgc, Muss.) 56. 335-344 10. Geier, A,, Beery, R.. Haimsohn. M. & Lunenfeld, B. ( 1987) J. Steroid Hiocliem. 28.47 1-478 I 1. Martin, P. M.. Berthois, Y. & Jensen, E. V. ( I98X) /'roc. NutI. A c ~ dS. C ~ .U.S.A. 85,2533-2537
Received 23 November 19x9
Effect of changes in 5' coding sequence on level of expression of ovine growth hormone cDNA in Escherichia coli attention. Poor bacterial expression o f such hormones has been markedly improved by manipulation o f the sequencc 5' or 3' to the initiation codon [e.g. 1-51. It has been suggested that the original poor expression in thesc systems is caused by the formation of secondary structures in the mRNA which inhibit the initiation of translation. In most cases the nature of such secondary structures remains unclear. To investigate this problem we have previously constructed a series of plasmids in which thc cDNA sequence
A. JAMIL SAMI, 0.CARY L WALLIS and M IC HA EL WA LLI S Biochemistry Luhorutory, School of Biological Sciences, Utiivemity of Sussex. Fulmer, Rrightori BNI SQG, U.K . Production of biologically active growth hormones (GH) by recombinant DNA technology is receiving considerable Abhreviations used: GH, growth hormone; o. ovine. Plasrnid
v
... ...
pOGHel04
ATGACCATCCCAGCCATG
pOGHelO5
ATCACTCCCAGCCATG
oOGHelO6
ATGACCAACTTCCCAGCCATG
pOGHelO7 pOGHelO8
oGH variant content (as % total cell protein)
Nucleotide seauence
v
15
v .... v ATGACCGACTTCCCAGCCATG . . . v ATGACGCACTTCCCAGCCATG
22
...
3
v
pOGHelO9
ATGACCATTCCCGGGGACTTCCCAGCCATG . . . .
10
pOGHel10
ATGACAATTCCCGGGC;ACTCCCAGCCATG
....
3
pOGHel11
ATGACCATGATTACAGCCATG .
pOGHell2
ATCACCA~ATTCCCGGCGACTTCCCAGCCATG
pOGHell3
ATGACCATGA~'AATTCCCGGGGACMCCCCAGCCATG. . . .
pOGHelO3
ATGACCATGATTACCAAT~AGCCATG.
pOGHell4
ATGACCAI'GATTACGAATTCGGACTTCCCAGCCATG
pOGHel01
v
v
..
11
v
,
...
..
v ... v ATGACCATGAI'TACGAATTCCCGCGACTCCCAGCCATG . . .
11 11 15 12 27
Fig. 1. Effect of nircleotide seqirerice in expression plasmids on level of expression of oGH vuriunts The sequence of bases 3' to the initiation codon and the oGH content of induced bacteria containing the corresponding plasmid were determined as described in the text. Bases in italics are derived from pUC8 and the remainder are from the oGH coding sequence; V shows the site of deletions; underlined bases show differences from the B-galactosidase sequence in the first eight bases. Data for pOGHe 10 1 are from Wallis & Wallis 16, 8). Vol. I8
