Bisphosphonates antagonise bone growth factors' effects on human breast cancer cells survival

Abstract

Bone tissue constitutes a fertile 'soil' for metastatic tumours, notably breast cancer. High concentrations of growth factors in bone matrix favour cancer cell proliferation and survival, and a vicious cycle settles between bone matrix, osteoclasts and cancer cells. Classically, bisphosphonates interrupt this vicious cycle by inhibiting osteoclast-mediated bone resorption. We and others recently reported that bisphosphonates can also induce human breast cancer cell death in vitro, which could contribute to their beneficial clinical effects. We hypothesised that bisphosphonates could inhibit the favourable effects of 'bone-derived' growth factors, and indeed found that bisphosphonates reduced or abolished the stimulatory effects of growth factors (IGFs, FGF-2) on MCF-7 and T47D cell proliferation and inhibited their protective effects on apoptotic cell death in vitro under serum-free conditions. This could happen through an interaction with growth factors' intracellular phosphorylation transduction pathways, such as ERK1/2-MAPK. In conclusion, we report that bisphosphonates antagonised the stimulatory effects of growth factors on human breast cancer cell survival and reduced their protective effects against apoptotic cell death. Bisphosphonates and growth factors thus appear to be concurrent compounds for tumour cell growth and survival in bone tissue. This could represent a new mechanism of action of bisphosphonates in their protective effects against breast cancer-induced osteolysis.

Figure 1

Modulation of breast cancer cell survival by growth factors. Breast cancer cells (○, MCF-7 and ▪, T47D) were incubated for 24 h under serum-free conditions, without or with increasing concentrations of IGF-I, IGF-II, FGF-2 or EGF. Cell viability was evaluated using the MTT test and expressed as treated over control ratios (means±s.e.m., n=15–20).

Figure 2

Bisphosphonates reduce growth factors' stimulatory effects on human breast cancer cell survival. MCF-7 cells (upper panel) and T47D cells (lower panel) were incubated for 24 h under serum-free conditions, in the presence of 10⁻⁶ M pamidronate (P), clodronate (C), ibandronate (I) or zoledronate (Z), and of IGF-I (10 ng ml⁻¹), IGF-II (100 ng ml⁻¹) or FGF-2 (5 ng ml⁻¹). Cell viability was evaluated by the MTT test and expressed as the percentage of growth factors' effects alone (means±s.e.m., n=18–20). ^*:P<0.05 vs growth factor alone.

Figure 3

Modulations of effector caspases activity in MCF-7 cells. Subconfluent MCF-7 cells were incubated in the presence of 10⁻⁶ M bisphosphonates and IGF-I (10 ng ml⁻¹), IGF-II (100 ng ml⁻¹) or FGF-2 (5 ng ml⁻¹) for up to 48 h. Effector caspases (caspase -6, -7) activity was determined at different time points using a synthetic fluorogenic substrate (DEVD-AMC, see Materials and methods). Results are shown as treated over control ratios (n=3).

Figure 4

Modulations of the bax/bcl-2 ratio in MCF-7 cells. Subconfluent MCF-7 cells were incubated for 24 h, under serum-free conditions, in the absence or presence of 10⁻⁶ M pamidronate (P), clodronate (C), ibandronate (I), zoledronate (Z), 10 ng ml⁻¹ IGF-I or 5 ng ml⁻¹ FGF-2. Bax and bcl-2 expression levels were determined by Western blot and corrected for β-actin levels. Results are expressed as means±s.e.m. of the bax/bcl-2 ratio (n=4–10). ^*:P<0.05 vs control. (A) Blots and derived ratios under bisphosphonates treatment. (B) Blots and derived ratios under growth factors treatments. (C) Relative intensity of signals under combinations of growth factors (IGF-I or FGF-2) and bisphosphonates. a: P<0.05 vs control; b: P<0.05 vs growth factor alone.

Figure 5

Modulation of the ERK-MAPK phosphorylation status in MCF-7 cells. (A) MCF-7 cells were incubated for 10 min with bisphosphonates (10⁻⁶ M) and/or FGF-2 (5 ng ml⁻¹) in phenol red-free and serum-free medium. Cell lysates were analysed by Western blot for phospho-p42/p44 and total p42 expression as described in Materials and methods section. (B) Relative intensity of signals, after scan and software analyses (ImageQuant). Results are expressed as percentages of controls. PAM=pamidronate; CLOD=clodronate; IBAN=ibandronate and ZOLE=zoledronate.