Use of synthetic isoprenoids to target protein prenylation and Rho GTPases in breast cancer invasion

Abstract

Dysregulation of Ras and Rho family small GTPases drives the invasion and metastasis of multiple cancers. For their biological functions, these GTPases require proper subcellular localization to cellular membranes, which is regulated by a series of post-translational modifications that result in either farnesylation or geranylgeranylation of the C-terminal CAAX motif. This concept provided the rationale for targeting farnesyltransferase (FTase) and geranylgeranyltransferases (GGTase) for cancer treatment. However, the resulting prenyl transferase inhibitors have not performed well in the clinic due to issues with alternative prenylation and toxicity. As an alternative, we have developed a unique class of potential anti-cancer therapeutics called Prenyl Function Inhibitors (PFIs), which are farnesol or geranyl-geraniol analogs that act as alternate substrates for FTase or GGTase. Here, we test the ability of our lead PFIs, anilinogeraniol (AGOH) and anilinofarnesol (AFOH), to block the invasion of breast cancer cells. We found that AGOH treatment effectively decreased invasion of MDA-MB-231 cells in a two-dimensional (2D) invasion assay at 100 µM while it blocked invasive growth in three-dimensional (3D) culture model at as little as 20 µM. Notably, the effect of AGOH on 3D invasive growth was phenocopied by electroporation of cells with C3 exotransferase. To determine if RhoA and RhoC were direct targets of AGOH, we performed Rho activity assays in MDA-MB-231 and MDA-MB-468 cells and found that AGOH blocked RhoA and RhoC activation in response to LPA and EGF stimulation. Notably, the geranylgeraniol analog AFOH was more potent than AGOH in inhibiting RhoA and RhoC activation and invasive growth. Interestingly, neither AGOH nor AFOH impacted 3D growth of MCF10A cells. Collectively, this study demonstrates that AGOH and AFOH dramatically inhibit breast cancer invasion, at least in part by blocking Rho function, thus, suggesting that targeting prenylation by using PFIs may offer a promising mechanism for treatment of invasive breast cancer.

Figure 1. Prenyl Function Inhibitors.

Structure relationship between natural isoprenoids (A) and lead PFIs (B).

Figure 2. AGOH inhibits MDA-MB-231 cells invasion.

(A) MDA-MB-231 cells were treated with AGOH at the indicated concentration and then assessed for Matrigel invasion toward the combination of 5 ng/ml EGF and 250 ng/ml insulin. (B) Cells treated with AGOH for 3 days were harvested in RIPA buffer, immunoblotted with polyAG-antibody. Asterisk (*) symbolizes a p value <0.05. These results are representative from at least three separate experiments.

Figure 3. AGOH inhibits 3D invasive growth of MDA-MB-231 cells.

MDA-MB-231 cells were seeded in Matrigel and treated with AGOH at the indicated concentration. After culturing for 8 days, phase contrast images were taken from randomly chosen fields (A, C, E) or Matrigel containing colonies were fixed and immunostained for F-actin (phalloidin, red) and nuclei (DAPI, blue) (B, D). The representative images from three separate experiments are shown. Scale bars represent 50 µm.

Figure 4. Rho GTPases are required for the invasive growth of MDA-MB-231 cells.

Cells were electroporated with 5 µg GST or GST-C3 exotransferase, then seeded in 3D Matrigel and imaged (A) as described in Figure 3, or assessed for RhoA (B) and Rac1 (C) activities as described in the Methods section.

Figure 5. AGOH blocks RhoA and RhoC activation in response to LPA.

MDA-MB-231 cells (A and E) or MDA-MB-468 cells (B) were treated with 100 µM AGOH for 3 days, plated on collagen I coated dishes in the presence of PFI and then treated with 100 nM LPA or 5 ng/ml EGF as indicated for 5 min before harvesting for RhoA (A and B, quantified in C and D) or RhoC (E, quantified in F) activity assay. Rho activity assays are representative from at least three separate experiments.

Figure 6. AFOH blocks RhoA and RhoC activation in response to LPA or EGF.

MDA-MB-231 cells (A, C, E and F) or MDA-MB-468 cells (B and D) were treated with AFOH at the indicated concentration for 3 days, plated on collagen I coated dishes in the presence of AFOH and then treated with 100 nM LPA (A and E, quantified in C and F) or 5 ng/ml EGF (B, quantified in D) for 5 min prior to harvesting for RhoA (A–D) or RhoC (E, F) activity assays. Rho activity assays are representative from at least three separate experiments.

Figure 7. PFIs block p-MLC downstream of Rho signaling in MDA-MB-231 cells.

MDA-MB-231 cells were treated with 100 µM AGOH and 20 µM AFOH for 3 days, plated on collagen I coated dishes, and then treated with 100 nM LPA for 5 min. Cell lysates were immunoblotted with p-MLC. Total MLC-2 and β-actin serve as the loading controls.

Figure 8. AFOH blocks 3D invasive growth of MDA-MB-231 cells.

Cells were seeded in growth factor reduced Matrigel, treated with DMSO (A, B) or 5 µM AFOH (C, D) for 10 days. Then phase contrast images were taken from randomly chosen fields (A, C). Colonies grown in Matrigel were smeared onto slides, immunostained for F-actin (red) and nuclei (blue), and then imaged by confocal microscopy (B, D). The representative images from three experiments are shown.

Figure 9. AGOH and AFOH do not affect 3D growth of MCF10A cells.

MCF10A were seeded in Matrigel and treated with AGOH and AFOH at the indicated concentrations for 14 days. Then the phase contrast images (A) were taken from randomly chosen fields. Matrigel-containing colonies were smeared and immunostained for integrin α6 (green), E-cadherin (red) and nuclei (blue) (B). Diameter of 60–80 individual colonies was measured and averaged in each condition (C). Cells were treated with AGOH for 3 days, harvested in RIPA buffer, and immunoblotted with polyAG antibody (D). Representative images from at least three separate experiments are shown. Scale bars in phase contrast equal 50 µm and in confocal images represent 20 µm.