Bioluminescence Imaging of Prenylation Inhibition–Letter

We read with great interest the recent article by Chinault and colleagues (1), providing evidence that aminobisphosphonate zoledronic acid (ZOL) does not exert a direct antitumor activity against breast cancer cells in vivo. There is extensive preclinical evidence suggesting that aminobisphosphonates exhibit antitumor effects in addition to their therapeutic activity in preserving bone tissue. The underlying inhibitory mechanisms of aminobisphosphonates against tumor cells are primarily through the blockade of the enzyme farnesyl pyrophosphate synthase (FPPS) in the mevalonate pathway, which prevents prenylation of signaling proteins (e.g., Rho, Ras, Cdc42), leading to the inhibition of tumor cell adhesion, migration, invasion, and proliferation in vitro (2). Whether aminobisphosphonates exhibit direct and/or indirect antitumor effects in animal models remains a matter of intense discussion. Here, the authors used a bioluminescence-based imaging reporter system that is inducible by prenylation inhibitors, such as aminobisphosphonates, and they showed that the treatment of animals with ZOL fails to induce the reporter in human MDA-MB-231 breast tumor xenografts, indicating that the aminobisphosphonate does not directly target cancer cells in vivo (1). However, we would like to point out an aspect of tumor xenograft experiments not adequately addressed by Chinault and colleagues (1), which deserves a remark.

The inhibition of FPPS activity by aminobisphosphonates leads to the intracellular accumulation of IPP/ApppI mevalonate metabolites (2). We recently provided in vivo evidence that ZOL (administered at a dose compatible with clinical dosing regimens) induces accumulation of IPP/ApppI in human breast tumors implanted subcutaneously in animals and that human Vγ9Vδ2 T cells (a subset of human T cells) infiltrate and inhibit growth of these tumors producing high IPP/ApppI levels, but not those expressing low IPP/ApppI levels (3). IPP and ApppI are recognized by human Vγ9Vδ2 T cells as tumor phosphoantigens. Importantly, we showed that estrogen receptor–positive breast tumors (T47D, MCF-7) were more likely to produce IPP/ApppI after ZOL treatment because of a higher cellular uptake of ZOL and a higher activity of the mevalonate pathway compared with estrogen receptor–negative breast tumors (MDA-MB-231, MDA-MB-435; ref. 3). We believe these findings (3) may likely explain results obtained by Chinault and colleagues (1). It would be therefore very interesting to determine whether ZOL induces bioluminescence-based reporter activity in estrogen receptor–positive breast cancer cells in vivo.

See the Response, p. 6078