Reevaluation of the proposed autocrine proliferative function of prolactin in breast cancer

Abstract

The pituitary hormone prolactin (PRL) has been implicated in tumourigenesis. Expression of PRL and its receptor (PRLR) was reported in human breast epithelium and breast cancer cells. It was suggested that PRL may act as an autocrine/paracrine growth factor. Here, we addressed the role of locally synthesised PRL in breast cancer. We analysed the expression of PRL in human breast cancer tumours using qPCR analysis and in situ hybridization (ISH). PRL mRNA expression was very low or undetectable in the majority of samples in three cDNA arrays representing samples from 144 breast cancer patients and in 13 of 14 breast cancer cell lines when analysed by qPCR. In accordance, PRL expression did not reach detectable levels in any of the 19 human breast carcinomas or 5 cell lines, which were analysed using a validated ISH protocol. Two T47D-derived breast cancer cell lines were stably transfected with PRL-expressing constructs. Conditioned medium from the T47D/PRL clones promoted proliferation of lactogen-dependent Nb2 cells and control T47D cells. Surprisingly, the PRL-producing clones themselves displayed a lower proliferation rate as compared to the control cells. Their PRLR protein level was reduced and the cells were no longer responsive to exogenous recombinant PRL. Taken together, these data strongly indicate that autocrine PRL signalling is unlikely to be a general mechanism promoting tumour growth in breast cancer patients.

Fig. 1

Analysis of PRL mRNA expression in the pituitary gland, placenta, breast carcinoma and adjacent normal breast tissue by ISH. Bright field and dark field pictures are shown. A 507 bp PRL cRNA antisense probe (AS PRL) was hybridized onto sections of paraffin-embedded anterior pituitary, placental decidua, invasive ductal carcinoma of the breast and adjacent normal breast epithelium tissues as indicated. Additional adjacent sections were hybridized with PRL sense probe (S PRL) as a negative control. Signal is visualised under dark field as white grains. Bright field represents the haematoxylin and eosin stained sections after hybridization with PRL antisense probe. Under bright field ISH signal is visualised as black grains. Sections of anterior pituitary tissue were developed after 2 weeks of exposure whereas the other tissue sections were developed after 8 weeks of exposure. Representative pictures are shown at both low and high magnification (left and right panel, respectively). Scale bars = 100 μm

Fig. 2

Establishment of T47D/PRL clones with high levels of ectopic PRL expression a ISH analysis of PRL mRNA expression in wild type and PRL-transfected cell lines. Bright field (BF) and dark field (DF) pictures are shown. Sections of paraffin-embedded cells were hybridized with an antisense cRNA probe (507 bp) complementary to the PRL mRNA (AS PRL). Signal is visualised under DF as white grains. BF represents the haematoxylin and eosin stained sections after hybridization with AS PRL probe to show morphology of the cells. Under BF ISH signal is visualised as black grains. Lack of detectable hybridization signal with sense probe (S PRL) indicates specificity of signals obtained with the PRL AS probe. Sections were developed after 2 weeks of exposure. Scale bars = 100 μm. b Analysis of the PRL protein expression. Cells were grown to 70–80 % confluence in 10 % FCS medium and then starved for 24 h in serum-free DMEM medium. 20 μg of cellular protein extracts were analysed by Western blotting using a PRL-specific antibody (top panel); β-actin was used as protein loading control (lower panel). Supernatants were analysed for secreted and active PRL using ELISA and Nb2 proliferation, respectively. The PRL level is stated as nmoles/1 mio cells/24 h. b.d. below detection limit. c Transwell co-culture of T47D and T47D/PRL cells. T47D Tet-On cells were seeded in 24-well multi-dishes in 10 % CSS medium. After 24 h of plating, inserts with either T47D Tet-On or T47D/pTRE-PRL Cl.4 cells were placed on top of T47D Tet-On cells. After three days of culture, inserts were discarded. The PRL level in the medium was analysed by ELISA, while proliferation of T47D Tet-On cells was measured using ³H-thymidine incorporation. Mean ± SD of four independent wells are shown

Fig. 3

Ectopically expressed PRL reduces proliferation of T47D cells a, b Growth of T47D/PRL clones. Cells were seeded with a density of 4.8 × 10⁴ cells/well in a 24-well multi-dish in 10 % FCS medium (a) or 10 % CSS medium (b). Cell numbers were determined at the indicated days of culture. Mean ± SD of three independent wells are shown; c, d Proliferation of T47D/PRL clones. Cells were plated with a density of 1.2 × 10⁴ cells/well in a 96-well multi-dish in 10 % FCS medium, 10 % CSS medium or serum-free DMEM medium. Cell proliferation was estimated after three days of culture by measuring DNA synthesis using ³H-thymidine incorporation. Mean ± SD of six independent wells are shown; statistics is shown: ***p < 0.001; **p < 0.05; *p < 0.01; ns, p > 0.01 (t test). e Cell cycle analysis of T47D/PRL clones. Cells (6 × 10⁵) were plated in a 6-well multi-dish in 10 % FCS medium and cultured for three days. DNA of cells was stained with propidum iodide, analysed by flow cytometry and quantified using ModFit. Debris and doublets were excluded by FSC/SSC gating. G₀/G₁ and G₂-M populations are illustrated in red; the dashed grey field is the S phase. The DNA content is shown in percent ± SD of three independent wells; f Cyclin D1 and Bcl-2 protein expression in T47D/PRL clones. Cells were cultured for five days in 10 % FCS medium till ~80 % confluence. 20 μg of cellular protein were analysed by Western blotting using Cyclin D1 and Bcl-2 specific antibodies. β-actin was used as a protein loading control. The signals of Cyclin D1 and Bcl-2 were quantified and normalised by the β-actin signals. The numbers represent percentage of the respective control T47D cell line values. A representative of three independent experiments is shown

Fig. 4

Ectopically expressed PRL down-regulates PRLR expression and responsiveness to exogenous rPRL stimulation a PRLR protein expression in T47D/PRL cells. Cells were harvested at 70–80 % confluence and 40 μg protein extracts were analysed by Western blotting using an anti-human PRLR antibody. β-actin was used as protein loading control. The signals of PRLR were quantified and normalised by the β-actin signals. The numbers represent a percentage of the respective control T47D cell line. Expression of PRLR on the surface of the cells was analysed by flow cytometry (FACS). Cells were stained with an anti-human PRLR mAb followed by anti-mouse IgG-APC. The median fluorescent intensities (MFI) are shown. b Cell signalling induced by rPRL. Cells were grown to 70–80 % confluence in 10 % FCS medium and starved in serum-free DMEM medium for 24 h prior to treatment with 20 nM rPRL for 30 min. 20 μg of protein extracts were analysed for activation of the ERK1/2 pathway by Western blotting using an anti-pERK1/2 antibody. Membranes were stripped and sequentially re-probed with antibodies detecting total ERK1/2 protein and β-actin as loading controls. c Proliferation of T47D/PRL cells upon E2 and PRL exposure. Cells were seeded with a density of 1.2 × 10⁴ cells/well in 96-well multi-dishes in 10 % CSS medium. Next day, medium was renewed and cells were cultured for three days in absence or presence of 1 nM 17β-estradiol (E2) and 20 nM rPRL as indicated. Cell proliferation was estimated by measuring DNA synthesis using ³H-thymidine incorporation and stated in percent of 10 % CSS control for each cell line. Mean ± SD of six independent wells are shown for a representative of three independent experiments. Statistics is shown: ***p < 0.001; **p < 0.05; *p < 0.01; ns, p > 0.01 (t test)