Zoledronic acid treatment impairs protein geranyl-geranylation for biological effects in prostatic cells

Abstract

Background: Nitrogen-containing bisphosphonates (N-BPs) have been designed to inhibit osteoclast-mediated bone resorption. However, it is now accepted that part of their anti-tumor activities is related to interference with the mevalonate pathway.

Methods: We investigated the effects of zoledronic acid (ZOL), on cell proliferation and protein isoprenylation in two tumoral (LnCAP, PC-3,), and one normal established (PNT1-A) prostatic cell line. To assess if inhibition of geranyl-geranylation by ZOL impairs the biological activity of RhoA GTPase, we studied the LPA-induced formation of stress fibers. The inhibitory effect of ZOL on geranyl geranyl transferase I was checked biochemically. Activity of ZOL on cholesterol biosynthesis was determined by measuring the incorporation of 14C mevalonate in cholesterol.

Results: ZOL induced dose-dependent inhibition of proliferation of all the three cell lines although it appeared more efficient on the untransformed PNT1A. Whatever the cell line, 20 microM ZOL-induced inhibition was reversed by geranyl-geraniol (GGOH) but neither by farnesol nor mevalonate. After 48 hours treatment of cells with 20 microM ZOL, geranyl-geranylation of Rap1A was abolished whereas farnesylation of HDJ-2 was unaffected. Inhibition of Rap1A geranyl-geranylation by ZOL was rescued by GGOH and not by FOH. Indeed, as observed with treatment by a geranyl-geranyl transferase inhibitor, treatment of PNT1-A cells with 20 microM ZOL prevented the LPA-induced formation of stress fibers. We checked that in vitro ZOL did not inhibit geranyl-geranyl-transferase I. ZOL strongly inhibited cholesterol biosynthesis up to 24 hours but at 48 hours 90% of this biosynthesis was rescued.

Conclusion: Although zoledronic acid is currently the most efficient bisphosphonate in metastatic prostate cancer management, its mechanism of action in prostatic cells remains unclear. We suggest in this work that although in first intention ZOL inhibits FPPsynthase its main biological actitivity is directed against protein Geranylgeranylation.

Figure 1

Effect of zoledronate on proliferation of PC-3, PNT1-A and LNCaP. At D0, the cells were seeded into 96-well plates to obtain 80% confluence at D5 in control wells. At D1 and D3, cells were treated by vehicle or increasing doses of zoledronate (ZOL: 5, 10, 15, 20 μM). At D5, the cells were fixed with TCA and stained with 0.4% sulforhodamine. Staining intensity was quantified at 540 nm. Results are expressed as the ratio OD_treatedcells/OD_untreated of three independent assays each performed six times. Error bars indicate inter-assay mean ± 1 SD. * indicates a significant difference versus non-treated cells (p < 0.01). PC3 PNT1A LNCaP

Figure 2

Effect of mevalonate-derived products on 20 μM zoledronate-induced inhibition of prostatic cell proliferation. At D0, cells were seeded into 96-well plates to obtain 80% confluence at D5 in control wells. At D1, and D3, cells were treated by vehicle or 20 μM zoledronate (ZOL) in the presence or absence of intermediates of the mevalonate pathway (GGOH = geranyl-geraniol; MVA = mevalonate; FOH = farnesol; SQUA = squalene) at the indicated doses. Results are are expressed as the ratio OD_{treated cells}/OD_untreated of three independent assays each performed six times. Error bars indicate inter-assay mean ± 1 SD. * indicates a significant difference versus ZOL-treated cells (p < 0.01). PC-3 PNT1A LNCaP

Figure 3

Effect of zoledronate on farnesylation status of HDJ2. Prostatic cells where treated for 48 h with ZOL (20 μM), lysed in hot (90°C) SDS (1%) and analyzed by Western blot.

Figure 4

Effect of zoledronate on GGPP/GGOH incorporation into proteins in vitro (A) and in vivo (B). A: Geranyl-geranyl transferase I (GGTase I) was quantified using an in vitro test based on [³H]-prenyl pyrophosphate (0.5 μM, 8–10 Ci/mmol) incorporation into a mutant form of H-ras with a geranyl-geranylation CAXX box. The level of prenylation is expressed as a percentage of maximum incorporation of [³H]-prenyl, as determined by allowing the uninhibited reaction to go to completion. B: Western-blot analysis; PC-3 cells are treated by vehicle, ZOL 20 μM; ZOL 20 μM + FOH 10 μM; ZOL 20 μM + GGOH 10 μM.

Figure 5

Effect of zoledronate on LPA-induced stress fiber induction, a cellular effect dependent on RhoA geranylgeranylation. On D₀, PNT1-A cells were seeded onto glass coverslips in 6-well plates to obtain 60% confluence on D₃. On D₁, the cells were serum-starved till fixation on D₃. After treatment by the indicated drugs then stimulation by LPA (20 μM, 5 hours on D₃), actin fibers were visualized by tetramethylrhodamine isothiocyanate-labeled phalloidin. Cells are viewed on a Zeiss Axiophot microscope (X 630), and pictures are taken with a Princeton camera.

Figure 6

Biosynthesis of [¹⁴C]-cholesterol in PC3 cells under ZOL treatments. Cells were treated with ZOL (20 μM) for 2, 22 or 46 hours and [¹⁴C]-mevalolactone (2.25 μCi/ml) was added to the medium for 2 hours more. Sterols were separated by thin layer chromatography (silica gel F/ethyl acetate). [¹⁴C]-Cholesterol was revealed and quantified by autoradiography with Phophorimager^® (Molecular Dynamics).