Fig 1.
VEGFq forms a quadruplex structure in physiological solution and inhibits proliferation of lung cancer cells.
(A) Nucleotide sequence of VEGFq and MutVEGF oligonucleotides. (B) Circular Dichrosim spectroscopy of VEGFq and MutVEGF oligonucleotides. Peak absorbance at 260 nm and trough absorbance at 240nm are indicative of G-quadruplex formation for VEGFq, compared to the MutVEGF sequence, which did not form a quadruplex. (C) Dose response to VEGFq or MutVEGF in A549 cells after 144 h by MTT assay. Controls include untreated cells or cells treated with a quadruplex-forming oligonucleotide targeted to the HIF-1α promoter (HIF-1αq) and a mutated HIF-1α (MutHIF-1α) oligonucleotide which did not form quadruplex.
Fig 2.
VEGFq inhibits cell proliferation of NSCLC cells but not non-transformed cells in a concentration and time dependent manner.
Proliferation was measured by the MTT assay. (A) VEGFq, but not MutVEGFq inhibited the growth of A459 cells in a concentration and time dependent fashion. (B) VEGFq also inhibited the growth of the H1299 and H1944 NSCLC cell lines, H3255 squamous cell lung cancer cells, and the Calu-1 epidermoid carcinoma cell line, but had very little effect on the non-transformed foreskin fibroblast Hs27 cells and the lung epithelial cells, HPLD-1 measured at 144 h. Bars represent mean±SEM absorbance of three separate determinations. * indicates (p<0.05) compared to untreated cells.
Fig 3.
VEGFq inhibit cell growth but does not increase cell death in Lung cancer cells.
A549, H1299 and HPLD-1 were exposed to 10μM of VEGFq or MutVEGFq for 3, 4 or 5 days and then counted in Trypan blue. (A) Percent of cell growth compared to untreated, inset: total number of cells. (B) Percentage of cell death. Data represent the averaged cell counts for 3 separate experiments realized in duplicate +/- SEM, * for p<0.05.
Fig 4.
VEGFq induced inhibition of cell growth can be partially rescued by recombinant VEGF and the absence of cell cycle changes.
(A) The effect of treatment of A549 cells with recombinant VEGF (1–50 ng/ml) with and without VEGFq pretreatment on cell proliferation after 144h by MTT assay. Bars represent mean±SEM absorbance of three separate determinations. * indicates (p<0.05) compared to untreated cells. Results show that addition of recombinant VEGF to VEGFq-treated cells significantly diminishes the growth inhibitory effect of VEGFq treatment. (B) Flow cytometry analysis of cell cycle for A549 cells treated with VEGFq and MutVEGF. No changes in the cell cycle were noted.
Fig 5.
Cellular uptake and stability of VEGFq or MutVEGF in A549 cells.
(A) Uptake of the FITC-labeled VEGFq or MutVEGF oligonucleotides (10μ M) in A549 and Hs27 after 1h, 24h, or 72h determined by FACS analysis (B). Confocal microscopy analysis of A549 or non-transformed fibroblast Hs27 cells treated with 10μM FITC-labeled VEGFq or MutVEGF after 72hr. In microscopic analysis, cell nuclei were stained with DAPI. Note significant uptake of VEGFq compared to MutVEGF in A549 and Hs27 cells. (C-D) 32P-labeled VEGFq and MutVEGF sequences were incubated in DMEM media at 37°C with 10% FBS (C) or A549 S100 cell extract (D) for 0–72 h. Black arrows indicate intact ODNs and red arrows the denatured products. Greater serum and intracellular stability occurred with VEGFq compared to MutVEGF. Note immediate degradation of MutVEGF into a secondary product.
Fig 6.
VEGFq induces authophagic cell death in NSCLC cells.
Electron microscopy images of untreated A549 cells (A, B) and A549 cells treated with 10μM VEGFq for 96h (C, D). VEGFq treatment caused dense cytoplasmic accumulation of autophagic vacuoles compared to untreated cells. (original magnification 8800X-A, C; 3400X-B, D). (E) FACS analysis of LC3B staining (FITC) as an autophagosome marker, confirmed induction of autophagy. Treatment of A549 cells with VEGFq (green) for 96h increased LC3B expression compared to untreated (blue) and cells treated with MutVEGF (orange). Minimal staining was noted with secondary antibody only (teal), confirming specificity.
Fig 7.
VEGFq inhibits A549 cell invasion.
(A) Treatment of A549 cells with VEGFq (10 μM) significantly decreased cell invasion (Matrigel) after 24 h compared to untreated and cells treated with MutVEGF determined by Boyden chamber. No change in migration (control) was noted. (B) Invading cell counts were normalized to migrating cell counts to calculate a normalized invasion index. Bars represent mean±SEM absorbance of three separate determinations. * indicates (p<0.05) compared to untreated cells.
Fig 8.
VEGFq treatment decreases newly transcribed VEGF mRNA and protein expression and downstream ERK1/2 and AKT mediated signaling.
(A) Nuclear run-on VEGF mRNA after 72h and (B) VEGF protein expression after 96/144h exposure to 10μμM VEGFq or MutVEGF as determined by RT-PCR and Western blot analysis respectively. Note a 40% decrease in newly synthesized VEGF mRNA and a profound decrease in protein expression. Actin was used as the loading control for both RT-PCR and Western blot experiments. Bars represent mean±SEM absorbance of three separate determinations. * indicates (p<0.05) compared to untreated cells. Blots are a representation of three independent experiments. (C) Time course of ERK 1/2 / p-ERK 42/44 and (D) AKT / p-AKT Ser 473 protein expression after 24-144h of 10μM VEGFq or MutVEGF treatment as determined by Western analysis. Actin was used as the loading control. Note substantial loss of ERK 1/2 and AKT activity after VEGFq treatment. Blots are a representation of three independent experiments.
Fig 9.
VEGFq binds to its double stranded target sequence by strand invasion.
(A) Electrophoretic mobility shift assay (EMSA) of VEGFq binding to an 84bp duplex target sequence. Radiolabeled VEGFq oligonucleotide was added to an excess of target sequence. Two retarded bands were observed, both of which were abrogated by the addition of unlabeled VEGFq, indicating sequence specific binding. (B) EMSA representing the binding of VEGFq to the C-rich single strand of the target VEGF promoter sequence, as expected, no binding was noted to the G-rich strand suggesting the VEGFq binds to the C-rich strand by strand invasion, rather than to the G-rich strand by intermolecular G-quadruplex formation.
Fig 10.
Cartoon depicting the “strand invasion model” of VEGFq action in NSCLC.
VEGFq binds to the VEGF promoter and stabilize the G-quadruplex structure on the opposite stand, this results in downregulation of VEGF transcription. The reduced expression of VEGF failed to activate MAPK pathway, notably the activation of ERK and AKT that has been shown to inhibit autophagy (red arrows). Therefore, by altering the MAPK pathway, VEGFq inhibits proliferation and invasion and promotes autophagic cell death.