HIF-dependent antitumorigenic effect of antioxidants in vivo

Abstract

The antitumorigenic activity of antioxidants has been presumed to arise from their ability to squelch DNA damage and genomic instability mediated by reactive oxygen species (ROS). Here, we report that antioxidants inhibited three tumorigenic models in vivo. Inhibition of a MYC-dependent human B lymphoma model was unassociated with genomic instability but was linked to diminished hypoxia-inducible factor (HIF)-1 levels in a prolyl hydroxylase 2 and von Hippel-Lindau protein-dependent manner. Ectopic expression of an oxygen-independent, stabilized HIF-1 mutant rescued lymphoma xenografts from inhibition by two antioxidants: N-acetylcysteine and vitamin C. These findings challenge the paradigm that antioxidants diminish tumorigenesis primarily through decreasing DNA damage and mutations and provide significant support for a key antitumorigenic effect of diminishing HIF levels.

Figure 1

A. Representative human P493 B cell tumors with tetracycline repressible MYC marked with luciferase were imaged via a Xenogen instrument at day 10. Animals received normal water (Untreated) or water containing 40 mM N-acetylcysteine (NAC) or 0.01% (w/v) doxycycline (+DOX). B. Tumor volumes (Mean ±SD) measured by caliper at 2 weeks of a cohort of animals, including ones shown in 1A. Ten animals treated with both NAC and DOX displayed no tumors (data not shown). C. Representative MYC-inducible neonatal hepatocellular carcinomas in the absence of doxycycline (Untreated) are evident through gross hepatomegaly, which was not observed in either doxycycline (+DOX) or NAC treated 11 day-old animals. D. Representative hematoxylin and eosin stained histological sections of livers from untreated, DOX or NAC treated 8-day old neonates (originally 20x magnification). E. At least 10 animals in each untreated, doxycycline (DOX) or NAC treated groups are shown with the corresponding liver to body mass ratio (upper panel) or Myc mRNA level (lower panel). Mouse age is indicated in the bottom of the lower panel.

Figure 2

A. Immunoblots of HIF-1α protein levels in human P493 B cells grown in 20% or 1% oxygen in the presence or absence of 10 mM N-acetylcysteine (NAC) and the tubulin loading control are shown. Cells were treated with NAC or subjected to hypoxia for 24 hr before analysis. B. NAC (10 mM) treatment for 24 hr does not affect HIF-1α mRNA levels (Mean ± SD) in P493 cells. C. VEGF secreted into conditioned P493 media is markedly diminished by NAC, which does not directly inhibit the VEGF immunoassay in vitro. ELISA results (Mean ± SD) for duplicate measurements from four independent experiments are shown. D. Immunoblot of HIF-1α induced in hypoxic (1% oxygen for 8 hours) human prostate carcinoma PC3 cell line bearing an HRE-driven green fluorescent protein (GFP) reporter demonstrates the inhibitory effect of NAC on HIF-1α level. E. Flow cytometric analysis of GFP expression demonstrate induction of the HRE-responsive reporter in 1% oxygen that is diminished by 10 mM NAC, illustrating the functional effect of NAC on HIF-1α level and the reporter in the same cell line. Left panel. Dot plots represented by FL.1 and FL.2 parameters indicate the negative and positive PC3 cells expressing GFP. FL.1 represents a channel over the GFP emission spectrum peak, whereas FL.2 is a channel set over higher wavelengths providing an irrelevant parameter used to complete a display of the data in 2 dimensions. Right panel. Percent GFP fluorescence detected under 1% hypoxia over that at 20% oxygen in the presence or absence of 10 mM NAC.

Figure 3

A. Immunoblot of endogenous HIF-1α and the HIF-1 CA5 mutant in CA5 or control (Cont) hypoxic P493 cells untreated (Untreated) or treated with 10 mM N-acetylcysteine (NAC). Tubulin serves as a sample loading control. B. P493-6 cells were transfected with expression vectors encoding wild-type FLAG-HIF-1α or FLAG-HIF-1α (P402A/P564A) by using the Amaxa Nucleofector System. After 24h, cells were cultured with or without 10 mM NAC under 1% O₂ for an additional 24h. Cells were then lysed and immunoblotted with anti-FLAG antibody. Tubulin was also immunoblotted in the same membrane as a loading control. C. The P493 cells overexpressing CA5 were untreated or treated with siRNAs against prolyl hydroxylases PHD1 or PHD2. Immunoblots of HIF-1 (HIF-1α and CA5), PHD1, PHD2 and tubulin are shown. Note that CA5 is constitutively expressed while endogenous HIF-1α is expressed only in PHD2 siRNA treated cells in a manner that is not affected by NAC. D. RCC4 cells lacking VHL or reconstituted with VHL (RCC4 + VHL) were grown in 20% or 1% oxygen in the presence or absence of 10 mM NAC.

Figure 4

A. In vivo bioluminescent images at day 10 of representative CA5 overexpressing or control (Cont) P493 B tumors untreated (−) or treated (+) with NAC (40mM in drinking water). B. Volumes measured by caliper of CA5 overexpressing or control P493 B tumors untreated or treated with NAC (n = 10 for each group; mean ±SEM) are shown. C. Left panel. Immunoblot of HIF-1 demonstrates the hypoxic induction of endogenous HIF-1α in both control and CA5 overexpressing cells as compared with the constitutive expression of CA5. Right panel. Tumor cells in xenografts recovered from control (−) or NAC treated (+) mice display constitutive CA5 expression along with NAC-sensitive endogenous HIF-1α expression.

Figure 5

A. Representative in vivo bioluminescent images of SCID mice xenografts of P493 cells overexpressing the stabilized HIF-1 CA5 mutant or control transduced P493 cells. Animals were untreated or treated with ascorbate (ASC) (5g/liter in drinking water). B. Time-dependent tumor volumes determined by caliper in 4 groups of mice injected with P493 cells overexpressing CA5 or control cells and either untreated or treated with ascorbate (ASC). Data are shown as mean ± SEM. C. Immunoblot of HIF-1α in control P493 cells or P493 cells expressing CA5 untreated or treated with ascorbate (ASC). Endogenous HIF-1α is shown with ectopic HIF CA5 mutant.

Figure 6. HIF-1α levels are necessary for Myc-induced P493 tumorigenesis

A. Reduction of HIF-1α levels in P493 cells by stable shRNA expression as shown by immunoblotting. Control lysate (EV) is shown with two independent pools of P493 cells transduced with two different shRNAs (sh1718 and sh2265) targeting HIF-1α. B. In vivo bioluminescent images of tumors from control cells versus cells expressing HIF-1α shRNA. C. Tumor growth of control and HIF-1α expressing P493 cells as determined by caliper measurements (Mean + SEM; n = 5 for each group).

Figure 7. PC3 cells stably expressing the HIF-1α CA5 mutant are relatively resistant to NAC treatment

A. Immunoblot of lysates from hypoxic (1% oxygen) cells showing the expression of CA5 that is resistant to NAC-induced degradation. Cells were treated with 10 mM NAC for 24 hours. B. Tumor growth of control PC3 cells or PC3 cells stably expressing the HIF-1α CA5 mutant in response to NAC treatment (40 mM in drinking water). Data are shown as mean + SEM with n = 5 for each group.