Fig 2: MiR-99a targeted IGF1R in OSCC cells. (A) Comparison of nucleotides in the miR-99a seed sequence and the targets, mTOR (upper panel) and IGF1R (lower panel), within the 3'UTR among species. (B) Significant negative correlation between the expression of miR-99a and IGF1R protein in OSCC cell lines (r = -0.747, p = 0.0207) by Pearson correlation. Average amounts of IGF1R protein were determined in two independent blots. *p < 0.05. (C) No significant correlation between expression of miR-99a and mTOR protein in OSCC cell lines by Pearson correlation (r = -0.1164, p = 0.7655). Average amounts of mTOR proteins were determined in two independent blots. (D) Immunoblot analysis of IGF1R and mTOR expression in OEC-M1 and SCC15 cells with ectopic miR-99a expression (OEC-M1 miR-99a and SCC15 miR-99a) or non-silencing microRNA expressing controls (OEC-M1 NS and SCC15 NS). The protein levels were normalized against an internal control a-tubulin. Ratios were determined by dividing the normalized protein levels in miR-99a expressing cells with that in non-silencing microRNA expressing controls. The means in the lower graph were measured by averaging the ratios from independent blots. Bar, SE; *p < 0.05. (E) Ectopic miR-99a expression suppressed the expression of the Renilla luciferase reporter gene containing the wild type miR-99a binding sequence in IGF1R 3'UTR (IGF1R 3'UTR) in OEC-M1 miR-99a cells. The luciferase activity of the construct with the deleted miR-99a binding site for IGF1R 3'UTR (?IGF1R 3'UTR) was assigned as 1. Bar, SE; *p < 0.05.

Fig 3: Negative regulation of miR-99a expression by the IGF1R signaling pathway. (A) Downregulation of miR-99a in response to IGF1 treatment. After serum starvation, cells were treated with vehicle, 10% FBS or 10 nM IGF1. (B) Immunoblot analysis revealed increased levels of IGF1R protein following IGF1-induced downregulation of miR-99a. After serum starvation, cells were treated with vehicle, 10% FBS or 10 nM IGF1. The protein levels were normalized against an internal control a-tubulin. Ratios were determined by dividing the normalized protein levels in OEC-M1 cells with different conditions to that in complete medium (CM). The means in the lower panel were measured by averaging the ratios from independent blots. (C) The PI3K inhibitor LY294002 and (D) MAPK kinase inhibitor PD98059 suppressed the IGF1-induced downregulation of miR-99a. After serum starvation, cells were treated with vehicle, 10 nM IGF1, or combination of LY294002/PD98059 and IGF1. Expression of miR-99a was normalized against an endogenous control U6. The relative expression of miR-99a was determined by normalizing the expression of miR-99a in OEC-M1 cells in different conditions to that in cells in complete medium (CM). Bar, SE; *p < 0.05; **p < 0.01; ***p < 0.001. (E) Proposed model of the relationship between the activation of the IGF1R signaling pathway and miR-99a expression. The activation of the IGF1R signaling pathway initiated the inhibitory signal leading to decreased miR-99a expression. Following the suppression of the IGF1R signaling pathway by inhibitors, or its silencing by the removal of stimulation, we observed increased miR-99a expression because of the removal of the inhibitory signal. This reciprocal regulation of miR-99a and IGF1R signaling augmented the activation of the IGF1R signaling pathway in response to IGF1 stimulation and accelerated its inactivation following the removal of stimulation.

Fig 4: Effects of caveolin-1 depletion on intracellular signaling. ARPE-19 cells were transfected with Cav-1-siRNA or CTR-siRNA. After 72 h a batch of cells were serum starved overnight, then exposed in the presence (dark bars) or absence (white bars) of 100 nmol/L IGF-1 for 5 min. Then cells were lysed and immunoblotted with anti-phospho and anti-pan antibodies against IGF1R (a), AKT (b), and ERK1/2 (c). A representative Western blotting and quantification of densitometries of Western blot bands are shown. Data are expressed as mean ± SD of fold induction relative to ß-Actin (n = 3). *** p < 0.001 vs. the respective unstimulated batches of cells; §§§ p < 0.001 vs. IGF-1 and si-CTR + IGF-1.

Fig 5: mTOR inhibition activates the JAK/STAT pathway(A–D) Cells treated with/without rapamycin (20 ng/mL), Torin1 (250 nM), and RapaLink-1 (10 nM) for 24 h (A and C) or stable knockdown cells with mTOR shRNAs (B and D) were lysed, and immunoblot analysis was performed.(E–G) Cells were treated with Torin1 for 6 h, after which (E) JAK1/2 inhibitors (ruxolitinib: 2.5 µM, momelotinib: 5 µM), (F) IGF1R inhibitors, or (G) RTK inhibitors as described in Figure 3 were treated for an additional 18 h in the presence of Torin1. Cells were lysed, and immunoblot analysis was performed.(H) Cells were treated with/without mTOR inhibitors for 24 h, and immunoblot analysis was performed.(I and J) Cells were treated with Torin1 for 6 h, after which (I) IL-6ST inhibitor (SC144: 10 µM) or (J) STAT3 inhibitor (C188-9: 10 µM) was treated for an additional 18 h in the presence of Torin1. (I and J) Immunoblot analysis (I) or qRT-PCR (J) was performed. Data are the mean ± SD of three technical replicates. Statistical significance (*p < 0.05) was assessed by t test.See also Figure S4.