Cancer Letters, 66 (1992) 183 - 191 Elsevier Scientific Publishers Ireland Ltd.
Transcriptionally breast cancer
183
active non-ligand
binding oestrogen
Santo Marsigliantea, Simona Grecoa, Luciana Biscozzo”, Stewart Bakerc, Leo Giuseppeb, Gavin P. VinsonC and Carlo Storelli”
receptors
Barkerc,
in
Valerie
“Dipartimento di Biologia, Laboratorio Fisiologia Generale, ChiuersiM di Lecce, Via Provinciale per Monteroni, 73 100 Lecce, bLaboratorio Analisi Cliniche, Ospedale Multizona!e V. Fazzi, Lecce (Italy) and ‘Department of Biochemistry, Faculty of Basic Medical Sciences, Queen Mary and Westfield College, Mile End Road, London El 4NS (UK)
(Received 18 May 1992) (Revision received 20 July 1992) (Accepted 3 August 1992)
Summary Data were obtained on soluble oestrogen (ER) and progesterone (PR) receptors from 273 primary breast tumours, using an enzyme immunoassay (EIA) and ligand binding assay with dextran-coated charcoal (DCC) or isoelectric focussing separations. The p29 and total cathepsin D content was also assayed in the same samples. Tumours expressing ER (by either steroid binding assay or EIA) had higher levels of p29 than those which did not express the receptor (P < 0.0001). Moreover, tumours co-expressing ER, PR and p29 appeared to have higher levels of cathepsin D than those which were negative for at least one protein (P < 0.0001). Twenty out of the 273 human breast cancer samples, containing a fevel of ER positive by EIA, which did not bind labelled oestradiol, were identified; isoelectric focussing showed that such a receptor was also unable to bind hydroxytamoxifen. These tumours did not significantly differ from those Correspondence to: S. Marsigliante, Dipartimento di Biologia, Laboratorio Fisiologia Generale, Universita di Lecce, Via Provinciale per Monteroni, 73 100 Lecce, Italy.
0304-3835/92/$05.00 Printed and Published
0 1992 Elsevier Scientific Publishers in Ireland
in the whole population in their capacity to express ER (positive by EIA), PR, p29 and cathepsin D. It was concluded that the ElA can detect both a ligand binding ER and a receptor which is able to initiate PR transcription but does not bind radio-labelled ligands in vitro; such a receptor could have a bearing on the variability of tumour response to endocrine therapy.
Keywords: breast cancer; oestrogen; progesterone; ~29; cathepsin D; enzyme immunoassay; steroid binding assay Introduction
The presence in breast cancer cytosols of non-functional receptors, such as proteolytic fragments (meroreceptors) [27] and mutants lacking the regions A, B and even C [25] is well documented. These receptors bind exogenous labelled hormone and therefore are reported as ER positive by steroid binding assay (SBA), without being able to activate transcription. Moreover, a mutant receptor which is unable to bind ligand but induces transcription also exists [9 - 111. When such a Ireland Ltd.
184
receptor possesses the epitopes recognized by the monoclonal antibodies present in the Abbott ER-EIA kit, it can be routinely identified. The functionality of such receptor must be assessed if one wishes to correlate the response to endocrine manipulations with its presence. Such an evaluation can be obtained by assessing the co-expression of ER with oestrogen-induced gene products (PR, pS2, cath-D and growth factors) [ 1,X&18 - 24,281 or others ER-associated molecules such as the p29 protein [3 - 61. In the present study, a variety of techniques, all of which are capable of providing prognostic information, was performed on a group of 273 biopsies taken from patients with primary breast cancer who underwent mastectomy in the last 2 years. The purpose of this investigation addresses the question of whether the EIA can detect an oestrogen receptor which does not bind ligands. Furthermore, the biological and clinical implications of such ER are explored . Materials and Methods Tissue handling Human breast tumour tissue was obtained at operation. The tissue was placed on ice for periods of no more than lo- 15 min, until tuinour tissue could be histologically identified, excised and snap frozen in liquid nitrogen. Receptor assays For routine steroid binding assays (SBA) of ER and PR with the dextran coated charcoal (DCC) method procedures were used exactly as recommended by the EORTC [7]. The EIA for ER and PR were performed according to the instructions provided with the Abbott kit (Abbott Laboratories, WiesbadenDelkenheim, Germany). . In order to establish the possibility that nonoestradiol binding, immunoreactive ER, would bind tamoxifen, we used cytosols derived from ER negative by SBA (but positive by EIA) samples, which expressed PR (with concentrations ~10 fmol/mg protein by SBA and EIA).
Such cytosols were incubated with a saturating concentration (10 nM, final concentration) of labelled z-4-hydroxy-tamoxifen (OH-TAM) in the presence or absence of a 200-fold excess of cold OH-TAM for 20 h at 4OC. Free hormone was separated from bound hormone by incubation with dextran-coated charcoal at 4OC for 10 min, followed by centrifugation. An aliquot (270 ~1) of the radioactive supernatant (3 mg protein/ml) was used for isoelectric focussing (IEF). Aliquots were also treated with a solution of dextran-coated charcoal (0.5% charcoal, 0.05% dextran T70) for 30 min at 4OC to remove the endogenous steroids which could compete with the radiolabelled ligands. After removal of charcoal pellet, the ‘stripped’ cytosol was incubated with the labelled OHTAM and oestradiol and subjected to IEF. Isoelectric focussing (ZEF)of ER The IEF was performed exactly as described previously [ 171. p29 Protein assay A specific IRMA (ER-D5 IRMA kit by Amersham International plc, Amersham, Bucks, UK) was used and performed according to the instructions provided with the kit. Total cathepsin D assay The content of total cathepsin D (cath-D) was assayed by IRMA, according to the instructions provided with the Cis Biointernational Kit (Cis Biointernational Laboratories, Gif-sur-Yvette, France). Protein estimation Protein estimation was carried out using the method of Bradford [2]. Receptor concentrations were calculated as fmol ligand bound per mg of protein. Statistical analysis Statistical analysis of the correlations between concentration values of steroid receptors, p29 and cath-D was carried out using the Spearman’s rank correlation. Mann-WhitneyWilcoxon’s rank sum test (MWW) was used to
185
ER
SBA
(p29+) ER
EIA
p
Transcriptionally breast cancer
183
active non-ligand
binding oestrogen
Santo Marsigliantea, Simona Grecoa, Luciana Biscozzo”, Stewart Bakerc, Leo Giuseppeb, Gavin P. VinsonC and Carlo Storelli”
receptors
Barkerc,
in
Valerie
“Dipartimento di Biologia, Laboratorio Fisiologia Generale, ChiuersiM di Lecce, Via Provinciale per Monteroni, 73 100 Lecce, bLaboratorio Analisi Cliniche, Ospedale Multizona!e V. Fazzi, Lecce (Italy) and ‘Department of Biochemistry, Faculty of Basic Medical Sciences, Queen Mary and Westfield College, Mile End Road, London El 4NS (UK)
(Received 18 May 1992) (Revision received 20 July 1992) (Accepted 3 August 1992)
Summary Data were obtained on soluble oestrogen (ER) and progesterone (PR) receptors from 273 primary breast tumours, using an enzyme immunoassay (EIA) and ligand binding assay with dextran-coated charcoal (DCC) or isoelectric focussing separations. The p29 and total cathepsin D content was also assayed in the same samples. Tumours expressing ER (by either steroid binding assay or EIA) had higher levels of p29 than those which did not express the receptor (P < 0.0001). Moreover, tumours co-expressing ER, PR and p29 appeared to have higher levels of cathepsin D than those which were negative for at least one protein (P < 0.0001). Twenty out of the 273 human breast cancer samples, containing a fevel of ER positive by EIA, which did not bind labelled oestradiol, were identified; isoelectric focussing showed that such a receptor was also unable to bind hydroxytamoxifen. These tumours did not significantly differ from those Correspondence to: S. Marsigliante, Dipartimento di Biologia, Laboratorio Fisiologia Generale, Universita di Lecce, Via Provinciale per Monteroni, 73 100 Lecce, Italy.
0304-3835/92/$05.00 Printed and Published
0 1992 Elsevier Scientific Publishers in Ireland
in the whole population in their capacity to express ER (positive by EIA), PR, p29 and cathepsin D. It was concluded that the ElA can detect both a ligand binding ER and a receptor which is able to initiate PR transcription but does not bind radio-labelled ligands in vitro; such a receptor could have a bearing on the variability of tumour response to endocrine therapy.
Keywords: breast cancer; oestrogen; progesterone; ~29; cathepsin D; enzyme immunoassay; steroid binding assay Introduction
The presence in breast cancer cytosols of non-functional receptors, such as proteolytic fragments (meroreceptors) [27] and mutants lacking the regions A, B and even C [25] is well documented. These receptors bind exogenous labelled hormone and therefore are reported as ER positive by steroid binding assay (SBA), without being able to activate transcription. Moreover, a mutant receptor which is unable to bind ligand but induces transcription also exists [9 - 111. When such a Ireland Ltd.
184
receptor possesses the epitopes recognized by the monoclonal antibodies present in the Abbott ER-EIA kit, it can be routinely identified. The functionality of such receptor must be assessed if one wishes to correlate the response to endocrine manipulations with its presence. Such an evaluation can be obtained by assessing the co-expression of ER with oestrogen-induced gene products (PR, pS2, cath-D and growth factors) [ 1,X&18 - 24,281 or others ER-associated molecules such as the p29 protein [3 - 61. In the present study, a variety of techniques, all of which are capable of providing prognostic information, was performed on a group of 273 biopsies taken from patients with primary breast cancer who underwent mastectomy in the last 2 years. The purpose of this investigation addresses the question of whether the EIA can detect an oestrogen receptor which does not bind ligands. Furthermore, the biological and clinical implications of such ER are explored . Materials and Methods Tissue handling Human breast tumour tissue was obtained at operation. The tissue was placed on ice for periods of no more than lo- 15 min, until tuinour tissue could be histologically identified, excised and snap frozen in liquid nitrogen. Receptor assays For routine steroid binding assays (SBA) of ER and PR with the dextran coated charcoal (DCC) method procedures were used exactly as recommended by the EORTC [7]. The EIA for ER and PR were performed according to the instructions provided with the Abbott kit (Abbott Laboratories, WiesbadenDelkenheim, Germany). . In order to establish the possibility that nonoestradiol binding, immunoreactive ER, would bind tamoxifen, we used cytosols derived from ER negative by SBA (but positive by EIA) samples, which expressed PR (with concentrations ~10 fmol/mg protein by SBA and EIA).
Such cytosols were incubated with a saturating concentration (10 nM, final concentration) of labelled z-4-hydroxy-tamoxifen (OH-TAM) in the presence or absence of a 200-fold excess of cold OH-TAM for 20 h at 4OC. Free hormone was separated from bound hormone by incubation with dextran-coated charcoal at 4OC for 10 min, followed by centrifugation. An aliquot (270 ~1) of the radioactive supernatant (3 mg protein/ml) was used for isoelectric focussing (IEF). Aliquots were also treated with a solution of dextran-coated charcoal (0.5% charcoal, 0.05% dextran T70) for 30 min at 4OC to remove the endogenous steroids which could compete with the radiolabelled ligands. After removal of charcoal pellet, the ‘stripped’ cytosol was incubated with the labelled OHTAM and oestradiol and subjected to IEF. Isoelectric focussing (ZEF)of ER The IEF was performed exactly as described previously [ 171. p29 Protein assay A specific IRMA (ER-D5 IRMA kit by Amersham International plc, Amersham, Bucks, UK) was used and performed according to the instructions provided with the kit. Total cathepsin D assay The content of total cathepsin D (cath-D) was assayed by IRMA, according to the instructions provided with the Cis Biointernational Kit (Cis Biointernational Laboratories, Gif-sur-Yvette, France). Protein estimation Protein estimation was carried out using the method of Bradford [2]. Receptor concentrations were calculated as fmol ligand bound per mg of protein. Statistical analysis Statistical analysis of the correlations between concentration values of steroid receptors, p29 and cath-D was carried out using the Spearman’s rank correlation. Mann-WhitneyWilcoxon’s rank sum test (MWW) was used to
185
ER
SBA
(p29+) ER
EIA
p
