Caveolin-1 sensitizes rat pituitary adenoma GH3 cells to bromocriptine induced apoptosis

Abstract

Background: Prolactinoma is the most frequent pituitary tumor in humans. The dopamine D2 receptor agonist bromocriptine has been widely used clinically to treat human breast tumor and prolactinoma through inhibition of hyperprolactinemia and induction of tumor cell apoptosis, respectively, but the molecular mechanism of bromocriptine induction of pituitary tumor apoptosis remains unclear. Caveolin-1 is a membrane-anchored protein enriched on caveolae, inverted flask-shaped invaginations on plasma membranes where signal transduction molecules are concentrated. Currently, caveolin-1 is thought to be a negative regulator of cellular proliferation and an enhancer of apoptosis by blocking signal transduction between cell surface membrane receptors and intracellular signaling protein cascades. Rat pituitary adenoma GH3 cells, which express endogenous caveolin-1, exhibit increased apoptosis and shrinkage after exposure to bromocriptine. Hence, the GH3 cell line is an ideal model for studying the molecular action of bromocriptine on prolactinoma.

Results: The expression of endogenous caveolin-1 in GH3 cells was elevated after bromocriptine treatment. Transiently expressed mouse recombinant caveolin-1 induced apoptosis in GH3 cells by enhancing the activity of caspase 8. Significantly, caveolin-1 induction of GH3 cell apoptosis was sensitized by the administration of bromocriptine. Phosphorylation of caveolin-1 at tyrosine 14 was enhanced after bromocriptine treatment, suggesting that bromocriptine-induced phosphorylation of caveolin-1 may contribute to sensitization of apoptosis in GH3 cells exposed to bromocriptine.

Conclusion: Our results reveal that caveolin-1 increases sensitivity for apoptosis induction in pituitary adenoma GH3 cells and may contribute to tumor shrinkage after clinical bromocriptine treatment.

Figure 1

Bromocriptine enhances endogenous caveolin-1 mRNA expression in GH3 cells. (A) Panel a: Expression of caveolin-1 mRNA was enhanced after bromocriptine treatment. Panel b and c: Cellular proteins were extracted and Western blots performed 48 hours after recombinant caveolin-1 was transfected into GH3 cells (lane 2 and 5). Vehicle transfections (lane 1 and 4) were used as controls. Panel b is an immunoblot using rabbit anti-caveolin-1 antibody and panel c is an immunoblot using monoclonal anti-Myc antibody against Myc-tagged caveolin-1. Arrowhead indicates Myc-tagged caveolin-1. Arrow indicates endogenous caveolin-1. Protein extracted from A431 cells was used as an immunoblotting positive control. (B) Recombinant caveolin-1 expressed in A431 cells has identical perinuclear localization with endogenous caveolin-1. Myc-tagged caveolin-1 was transfected into A431 cells (a, c) for 48 hours. Non-transfected A431 cells were used to examine endogenous caveolin-1 localization (b, d). Cells were fixed and immunocytochemically stained with anti-caveolin-1 antibody then visualized by anti-rabbit IgG conjugated-FITC secondary antibody to detect exogenous (a) and endogenous (b) caveolin-1. Monoclonal anti-Myc antibody combined with anti-mouse IgG conjugated-Texas-Red was used to recognize recombinant caveolin-1 (c). Filamentary actin was stained with phalloidin conjugated-Texas-Red to observe cell morphology (d). Scale bar = 20 μm.

Figure 2

Overexpression of caveolin-1 induces nuclear condensation in GH3 cells. GH3 cells were transfected with either pcDNA4-caveolin-1 (A, C and E) or pcDNA4-EGFP (B, D and F). Immunocytochemical staining was carried out 48 hours after transfection. Phalloidin conjugated-Texas-Red (A and B) and Hoechst 33342 (E and F) dyes were used to stain F-actin and nuclei respectively. Cells expressing caveolin-1 (arrowhead) were visualized by anti-rabbit IgG conjugated-FITC secondary antibody (C). Caveolin-1 overexpressing cells exhibit nuclear condensation (E), while normal nuclei (arrow) were detected in cells over-expressing EGFP (F). Scale bar = 20 μm.

Figure 3

Over-expression of caveolin-1 induces apoptosis of GH3 cells. (A) GH3 cells were transfected with pcDNA4-Caveolin-1 for 48 hours then subjected to TUNEL assays. Cells transiently expressing exogenous caveolin-1 (indicated by arrowheads) showed shrunken morphologies (a and b) and nuclear fragmentation when stained with Hoechst 33342 dye (c) and were positively fluorescence-labeled by TUNEL assay (d). Cells directly treated with DNase I (e, f) or vehicle (g, h) were TUNEL positive (f) or negative (h), respectively. Nuclei stained with Hoechst 33342 dye are shown in c, e and g. (B) Caveolin-1 elicited apoptosis of GH3 cells. GH3 cells were transfected with pcDNA4-caveolin-1, pcDNA4-EGFP or vehicle (null treatment) and subjected to immunocytochemical staining 24 and 48 hours after transfection. Three hundred cells were randomly chosen and counted in each experiment (vehicle, caveolin-1 or EGFP) to determine the percentage with fragmented nuclei after Hoechst 33342 dye labeling. Apoptotic cells were measured by fluorescence labeling and data were expressed as mean ± standard deviation from n = 3 independent experiments (angular transformed for analysis, back-transformed for presentation). The standard deviations are too small to observe in the 48 hours data. **P < 0.01 and *P < 0.05 versus EGFP or vehicle experiment.

Figure 4

Caveolin-1 induces GH3 apoptosis via caspase 8 activity. Caspase inhibitors (50 μM) were added to either pcDNA4-Caveolin-1 or pcDNA4-DsRed-N1 (negative control) transfected GH3 cells and the proportion of apoptotic cells quantified 48 hours after transfection by immunocytochemical and TUNEL assays. Data were expressed as mean ± standard deviation from 3 independent experiments (angular transformed for analysis, back-transformed for presentation). Abbreviations on the x-axis: General caspase inhibitor (Z-VAD-fmk); a negative control inhibitor (Z-FA-fmk); specific caspase inhibitors for caspase 3 (Z-DEVE-fmk), caspase 8 (Z-IETD-fmk) or caspase 9 (Z-LEHO-fmk).

Figure 5

A combination of bromocriptine treatment and overexpression of Caveolin-1 enhances apoptosis of GH3 cells. (A) GH3 cells were transfected with caveolin-1 for 24 hours and then treated with or without 30 μM bromocriptine for another 12 hours. Non-transfected cells were treated for 12 hours with ethanol alone (vehicle) or bromocriptine. Cells expressing exogenous Caveolin-1 were immunostained with anti-Myc antibody and then detected with anti-mouse IgG Texas Red-conjugated secondary antibody. Three hundred Caveolin-1 expressing cells were counted in each experiment and apoptotic cells were determined by nuclear fragmentation after staining with Hoechst 33342 dye. Data were expressed as mean ± standard deviation from 3 independent experiments (angular transformed for analysis, back-transformed for presentation). (B) Phosphorylation of Tyr14 was enhanced when GH3 cells were exposed to bromocriptine. Cells were administered 30 μM bromocriptine as indicated, 24 hours after pcDNA4-caveolin-1 transfection. Cellular protein was extracted and separated on 12% SDS-PAGE then subjected to Western blotting. Anti-caveolin-1 or anti-phosphorylated caveolin-1 (Tyr¹⁴) specific antibodies were used to measure total and phosphorylated caveolin-1 respectively. Cellular protein extracts from NIH3T3 cells exposed to H₂O₂ were used as a positive control to detect phosphorylated caveolin-1. Dash (-) indicates vehicle treatment. The arrowhead points to endogenous caveolin-1 and the arrow marks recombinant caveolin-1. Abbreviation: Br, bromocriptine. *P < 0.05 versus experiment for caveolin-1 transient expression only or bromocriptine treatment only.