(W1030-10-55) Seeking Novel Drug Targets for Anti-Metastasis Treatment

Purpose: Chemotherapy options for cancer metastasis are largely lacking. Few drug targets have been identified due to our limited insights into the underlying molecular processes. Whereas early metastasis research focused on homing and invasion as key components, more recent in vivo studies have made it increasingly clear that disseminating cells linger in the circulation for extended periods of time. Their deadherent survival constitutes the main rate-limiting step in cancer dissemination. The detachment of healthy epithelial cells causes anoikis due to an impairment of energy production via ATP. In transformed epithelial cells, by contrast, Osteopontin splice variants support anchorage-independence through inducing the energy metabolism, which is essential for the spread of cancer cells. Over longer time frames, mitochondrial biogenesis becomes a prominent feature and further aids metastatic success.
We hypothesized that Osteopontin splice variants, which synergize to upregulate mitochondrial ATP generation in deadherent cells, also increase the mitochondrial mass via the same signaling cascades. The molecular mechanisms associated with this process will be candidate drug targets for anti-metastasis treatment.

Methods: Non-invasive MCF-7 breast tumor cells were stably transfected with full-length Osteopontin-a, alternatively spliced Osteopontin-c or vector control. Cells were grown in soft agar for 5 days (to measure mitochondria size) or 11 days (to measure colony size). The ability of cells to form colonies in soft agar correlates well with their ability to grow invasively in vivo. It does not only depend on the growth rate of the cells, but also on the expression of gene products for invasiveness and anti-anoikis. For analysis of the mitochondria, a prewarmed staining solution containing the MitoTracker^® probe was added to the plates followed by incubation for 30 minutes. The labeled mitochondria inside cell clusters in the top agar were photographed by confocal microscopy and their expansions were measured in ImageJ^®. For measuring the cluster size in soft agar, bright field pictures were taken in 5 random fields covering the dish. The colony areas were measured in ImageJ^®.

Results: Using the OPNa or OPNc transfectants (or vector) of MCF-7 cells, we measured the area sizes of fluorescently labeled mitochondria (as a substitute readout for mitochondrial mass, which is indicative of biogenesis) consecutive to 5 days of soft agar growth. Corresponding to the strong (OPNc) or intermediate (OPNa) induction of anchorage-independent growth, transfected OPNc significantly induced mitochondrial biogenesis, while transfected OPNa had an intermediate effect compared with the vector control. Osteopontin secretion is necessary and sufficient for stimulating mitochondrial biogenesis, and it occurs via autocrine binding to its receptor CD44v, not via engagement of Integrin receptors. We have identified the Peroxisome Proliferator-Activated Receptor γ Coactivator 1α (PPARGC1A) as a necessary intermediate in communicating the OPN effect. This signaling molecule promotes respiration and mitochondrial biogenesis. Consistently, strong PGC-1α expression is associated with distant metastasis and poor outcome in patients with invasive breast cancer. Further, peroxides have been widely described to mediate anchorage independence. The presence of the cell-permeable peroxide scavenger N-acetyl cysteine (NAC) was sufficient for suppressing the mitochondrial size induced by OPNc, partially suppress the response to OPNa, and not sufficient for affecting the mitochondrial size in vector control cells.

Conclusion: We have identified PPARG and peroxides as intermediates to be activated by Osteopontin under anchorage-deprived conditions. The resulting mitochondrial activation and biogenesis increase ATP production and enable anchorage independence (an essential feature of metastases). Treatment with inhibitors of Osteopontin or PPARG and/or administration of anti-oxidants may have therapeutic potential against cancer metastases.

References: [1] Z. Shi, B. Wang, T. Chihanga, M.A. Kennedy, G.F. Weber, Energy metabolism during anchorage-independence. Induction by osteopontin-c, PloS one, 9 (2014) e105675.
[2] V.S. LeBleu, J.T. O'Connell, K.N. Gonzalez Herrera, H. Wikman, K. Pantel, M.C. Haigis, F.M. de Carvalho, A. Damascena, L.T. Domingos Chinen, R.M. Rocha, J.M. Asara, R. Kalluri, PGC-1α mediates mitochondrial biogenesis and oxidative phosphorylation in cancer cells to promote metastasis, Nature cell biology, 16 (2014) 992-1003, 1001-1015.
[3] F. Hartung, Y. Wang, B. Aronow, G.F. Weber, A core program of gene expression characterizes cancer metastases, Oncotarget, 8 (2017) 102161-102175.

Acknowledgments: This research is supported by NIH grant CA224104 to GFW.

Figure 1. Autocrine OPN effectuates mitochondrial biogenesis via its receptor CD44v. A. OPN is necessary for mitochondrial biogenesis. Addition of anti-OPN antibody but not control Ig to deadherent cultures of MCF-7 transfectants every other day suppressed the mitochondrial size. B. OPN is sufficient for mitochondria biogenesis. Addition of 50%, 75%, 100% supernatants from vector, OPNa or OPNc to deadherent MCF-7 cultures increased the mitochondrial size. C. The OPN effect is exerted through CD44v. Antibodies to the CD44 variable domain, but not to the far N-terminus of CD44 or to Integrin β3, reversed the increase of mitochondrial size as well as soft agar colony formation of OPNa transfected MCF-7 cells. # = significant increase compared to vector control; * = significant reduction by treatment. Fluorescent intensity and colony size are expressed as relative units (control = 100 units). Sup = supernatant.



Figure 2. PGC-1α signaling is an essential mediator of the OPN effect on mitochondria. A. Mitochondrial size measurement of MCF-7 transfected cells under treatment with or without PGC-1α inhibitor SR-18292. The drug reduces the size of mitochondria in OPNa and OPNc transfected MCF-7 cells but not in vector transfectants. B. Western blot for PGC-1α and OPN expression levels. The PGC-1α inhibitor does not impact the PGC-1α expression levels in plated (bottom) or poly-HEMA (top) growth conditions.



Figure 3. Peroxide is an essential intermediary of the Osteopontin effect. Treatment of deadherent MCF-7 transfected cells with 2 mM NAC before assessment of colony size and mitochondria size. A. NAC suppresses the soft agar colony formation by OPNc transfected MCF-7 cells. B. NAC inhibits the mitochondrial size of OPNc cells.