Fig 2: Immunohistochemistry staining of xenografts on 12, 15, and 28 d after the tumor implantation. (A) Heterotransplantation tumors P4-pBIG2i and P4-pBIG2i-hIGFBP3 at 12, 15, and 28 d post-tumor implantation on IGFBP3 (100 ×), HIF-1a (40 × and 100 ×), HIF-2a (40 × and 100 ×), HO-1 (100 ×), and VHL (100 ×). P4-pBIG2i xenograft showed no IGFBP3 expressions, while P4-pBIG2i-hIGFBP3 showed a strong IGFBP3 expression after adding doxycycline (started on d 8). HIF-2a increased over time in P4-pBIG2i xenograft, but it only accumulated after 28 d in P4-pBIG2i-hIGFBP3 xenograft. Xenograft tumors without IGFBP3 expressed HIF-2a majorly, especially under a prolonged tumor growth in hypoxic environments. (B) Quantitative analysis of immunohistochemical staining. Images of IHC were taken by Motic Image Plus 3.0 software. The signals of IHC were quantified and analyzed by ImageJ 1.53 k software. Excel 2016 was used to generate charts. Photoshop CS2 version 9 was used to assemble the figure.

Fig 3: Clinical roles of HIF-1a and HIF-2a related to IGFBP3 expression in EOC. In an early stage of EOC, IGFBP3 expresses and inhibits proliferation and angiogenesis. The lack of blood vessels causes tumors to enter a hypoxic state, which stimulates cells to activate HIFs. HIF-1a is the major HIFs in this tumor stage. HIF-1a stimulates angiogenesis, proliferation, and induces IGFBP3 synthesis. The continuous expression of IGFBP3 prolongs the hypoxic state of the tumors by inhibiting tumor vasculogenesis. Prolonged hypoxia causes the promoter to be methylated and silence the expression of IGFBP3. As IGFBP3 decreases, the cells begin to proliferate and switch to activate HIF-2a majorly under hypoxic conditions. HIF-2a plays a significant role in cancer aggressiveness by regulating angiogenesis, invasion, and metastasis. Henceforth, the cancer cells mainly express HIF-2a under hypoxic stress. The decrease in IGFBP3 and the activation of HIF-2a instead of HIF-1a accelerate the progression in EOC. PowerPoint 2016 and Photoshop CS2 version 9 were used to generate figure.

Fig 4: IGFBP3 differentially regulated HIF-1a and HIF-2a synthesis in low and high invasion ability cells. Gene expression and regulation of IGFBP3, HIF-1a, and HIF-2a in cells cultured under normoxic and hypoxic conditions. (A) qPCR analyzed of IGFBP3, HIF-1a, and HIF-2a mRNA expressions in P0 and P4 cultured under normoxic and hypoxic conditions. GAPDH as the normalized control. P4 expressed higher HIF-1a and HIF-2a under normoxia, and P0 expressed lower HIF-2a compare with P4. (B) Luciferase promoter assays on the regulations of IGFBP3, HIF-1a, and HIF-2a promoters in P0 and P4 under normoxic and hypoxic conditions. The table below indicates the presence of endogenous or exogenous IGFBP3. In high-IGFBP3 expression P0, HIF-1a was more activated than HIF-2a at hypoxia. Both HIF-1a and HIF-2a were reduced after depletion of extracellular IGFBP3 by neutralizing antibodies (lanes 4 and 6). In low-IGFBP3 expression P4, both HIF-1a and HIF-2a activities increased under hypoxia. There were no changes in HIF-1a and HIF-2a expressions after adding recombinant hIGFBP3 (lanes 8 and 10). Each sample was assayed and analyzed repeated three times independently. (The error bars represent the SDs; **P < 0.001, ***P < 0.0005.). Excel 2016 was used to generate charts. Photoshop CS2 version 9 was used to assemble the figure.

Fig 5: Protein expression analysis of cell cultures under normoxic and hypoxic conditions. (A) Western blot analysis of IGFBP3, HIF-1a and HIF-2a, HO-1, VHL, and GAPDH expressions in P0, P4, P4 transfectants, and A549 under normoxic and hypoxic conditions. HSP90 was added as the control. (B) Western blot results were transformed into bar figures using Image Studio Lite version 5.2. Signal strengths were quantified and normalized with HSP90. P0 had a higher IGFBP3 expression than P4. In P0 and P4-I, IGFBP3 expression was increased under hypoxia. Accumulation of HIF-1a or HIF-2a correlated with IGFBP3 expressions: a higher IGFBP3 was with a stronger HIF-1a. Both HO-1 and VHL expressions were increased under hypoxia. Each sample was assayed in triplicates, and the experiment was repeated three times independently. (The error bars represent the SDs; **P < 0.001, ***P < 0.0005). Excel 2016 was used to generate charts. Photoshop CS2 version 9 was used to assemble the figure.