Potent Antitumor Effects of a Combination of Three Nutraceutical Compounds

Abstract

Head and neck squamous cell carcinoma (HNSCC) is associated with low survival, and the current aggressive therapies result in high morbidity. Nutraceuticals are dietary compounds with few side effects. However, limited antitumor efficacy has restricted their application for cancer therapy. Here, we examine combining nutraceuticals, establishing a combination therapy that is more potent than any singular component, and delineate the mechanism of action. Three formulations were tested: GZ17-S (combined plant extracts from Arum palaestinum, Peganum harmala and Curcuma longa); GZ17-05.00 (16 synthetic components of GZ17-S); and GZ17-6.02 (3 synthetic components of GZ17S; curcumin, harmine and isovanillin). We tested the formulations on HNSCC proliferation, migration, invasion, angiogenesis, macrophage viability and infiltration into the tumor and tumor apoptosis. GZ17-6.02, the most effective formulation, significantly reduced in vitro assessments of HNSCC progression. When combined with cisplatin, GZ17-6.02 enhanced anti-proliferative effects. Molecular signaling cascades inhibited by GZ17-6.02 include EGFR, ERK1/2, and AKT, and molecular docking analyses demonstrate GZ17-6.02 components bind at distinct binding sites. GZ17-6.02 significantly inhibited growth of HNSCC cell line, patient-derived xenografts, and murine syngeneic tumors in vivo (P < 0.001). We demonstrate GZ17-6.02 as a highly effective plant extract combination and pave the way for future clinical application in HNSCC.

Figure 1

GZ17 formulations are cytotoxic to HNSCC, and have a potentiated effect compared to individual components. (A) OSC19 (4 × 10³ cells/well in triplicate) were treated with various concentrations of GZ17-6.02, -5.0 and -S. Effective dose 50 (ED₅₀) was calculated with non-linear curve fit using GraphPad Prism software. Cumulative data represents three individual experimental repeats and error bars represent ± SEM. (B) OSC19 (4 × 10³ cells/well in triplicate) were treated with curcumin, harmine or isovanillin or combination of two components, each in a ratio representative of GZ17-6.02 in a final dose of 50 ug/mL for 48 h. (C) Glioblastoma (U87), and lung cancer lines (201T and A549) were treated with various concentrations of GZ17-6.02 to determine ED₅₀ concentration. (D) HNSCC cells (OSC19; 2 × 10⁵ cells) were treated with vehicle control or ED₅₀ concentrations of GZ17-6.02, -05.00 or –S for 72 h and analyzed by flow cytometry. The percentage of cells in various cell cycle stages is represented of each treatment of GZ17 formulation at ED₅₀ concentration. Graph represents cumulative results from three independent experiments. (E) Representative immunoblot of apoptotic markers cleaved-PARP and caspase-3. β-tubulin levels demonstrate equal loading of protein across lanes.

Figure 2

GZ17 formulations mitigate HNSCC invasion and migration, and angiogenesis. (A,B) HNSCC cells (OSC19; 2 × 10³ cells/ well, plated in triplicate) were treated with vehicle control or ED₅₀ concentrations of GZ17-6.02, -05.00 or -S. Cell migration and invasion was assessed at 24 h. The number of cells that (A) migrated or (B) invaded were counted and normalized to the cell viability. Percent migration or invasion relative to the vehicle control is depicted in the graphs. Cumulative data represents three individual experimental repeats and error bars represent ± SEM. (C,D,E) Glioblastoma (U87), and (D and E) lung cancer (201T and A549) were assessed for GZ17-6.02 inhibition of migration and invasion. Cumulative data represents three individual experimental repeats and error bars represent ± SEM. (F) GZ17-6.02 attenuates the angiogenic potential of HUVEC in vitro. HUVEC cells were treated with the ED₅₀ dose (derived from OSC19) of GZ17-6.02 and imaged 6 h after treatment and tubule formation was assessed. Total tube length analyzed using Pipeline software from 15 random fields from each repeat, and normalized to vehicle control treated cells (VC). Cumulative data represents three individual experimental repeats and error bars represent ± SEM. (G) GZ17-6.02 inhibits tumor-promoting macrophage survival. Macrophage cell line Thp1 were treated with 60 µg/ml GZ17-6.02 for 48 h. Viable cells were counted using trypan blue dye exclusion. Graph represents cumulative results from five independent experiments and error bars represent ± SEM.

Figure 3

GZ17 formulations modulate levels of several phospho-proteins in HNSCC. (A) HNSCC cells (OSC19; 2 × 10⁵ cells) were treated with vehicle control or ED₅₀ concentrations of GZ17-6.02, -05.00 or –S for 48 h. Representative dot-blot image from phospho-kinase array. (B) Densitometric analyses of dot-blot image signals from GZ17 treated lysates normalized to those from vehicle control and presented as fold change in protein levels. The graph represents cumulative data from two independent experiments. Error bars represent ± SEM. (C) HNSCC cells (OSC19; 2 × 10⁵ cells) were treated with ED₅₀ concentrations of GZ17-6.02, -05.00, –S or vehicle control for 72 h. Immunoblot was performed for phospho-ERK1/2 and total ERK1/2 as loading control. Image is representative of three independent experimental repeats. (D) Densitometric analyses of signals from immunoblots normalized to loading control and presented as fold change in protein levels relative to vehicle control treated cells. Error bars represent ± SEM.

Figure 4

Curcumin and harmine bind with high affinity to distinct sites on EGFR, ERK1 and Akt-1. Binding conformation of top ranked docked poses of (A,C,E) curcumin and (B,D,F) harmine on EGFR tyrosine kinase domain, ERK1 tyrosine kinase domain and Akt1 catalytic domain, respectively. Further details, including detailed labeling of interactive amino acid residues are provided in Supplemental Figs 3–6.

Figure 5

GZ17-6.02 in combination with cisplatin potentiates HNSCC cytotoxicity. HNSCC cells (A) OSC19, (B) HN5, and (C) UM-SCC-1; (2 × 10³ cells/well) in triplicate were treated with respective ED₅₀ concentrations of GZ17-6.02, cisplatin (4 µM), combination of both GZ17-6.02 and cisplatin, or vehicle control for 72 h. Cell survival was assessed by CyQUANT assay. Obtained values were normalized to vehicle control and depicted as fold change in cell survival. Cumulative data represents three individual experimental repeats plated in triplicate and error bars represent ± SEM.

Figure 6

GZ17-6.02 demonstrate significant reduction in HNSCC tumor growth in vivo. (A) HNSCC cells (OSC19; 1 × 10⁶ cells) were inoculated subcutaneously on the flanks of athymic nude-Foxn1^nu mice. Five mice per group were treated intratumorally with vehicle control (saline), or 15 mg/kg/d of GZ17-06.02, GZ17-5.0 or GZ17-S for three weeks. The graph depicts tumor volumes measured with a vernier caliper over the course of the experiment. Error bars represent ± SEM. (B) Immunocompetent SCC/vII tumor bearing C3H mice were treated with GZ17-6.02 (100 mg/kg/day in 1% carboxymethylcellulose (CMC) suspension) or VC (1% CMC) by oral gavage (N = 10/group). Graph depicts tumor volumes measured with a vernier caliper over the course of the experiment. (C) Patient-derived HNSCC tumor masses (35 mg/site) were implanted subcutaneously on both flanks of athymic nude-Foxn1^nu mice. Ten mice per group were treated by oral gavage with GZ17-6.02 (30 mg/kg/d for first 7 days, and dose increased to 50 mg/kg/d to improve antitumoral effect) or vehicle control (saline) for 19 days. The graph depicts fractional tumor volumes over the course of the experiment. (D) Representative immunoblot of patient-derived xenograft lysates demonstrates decrease in p-ERK1/2 levels. Densitometric analysis of p-ERK1/2 relative to density of loading control (β-tubulin) of GZ17-6.02 treated tumors (n = 8) compared relative to vehicle control (n = 8) were graphed. Error bars represent ± SEM.