Conceived and designed the experiments: HS WM CUL PMH. Performed the experiments: HS WM MFS. Analyzed the data: HS WM MFS CUL PKL JK PW PMH. Contributed reagents/materials/analysis tools: MFS JK CUL PKL MNS. Wrote the paper: PW PMH.
The authors have declared that no competing interests exist.
Oxygen serves as an essential factor for oxidative stress, and it has been shown to be a mutagen in bacteria. While it is well established that ambient oxygen can also cause genomic instability in cultured mammalian cells, its effect on
Oxygen is essential for aerobic life and its deficiency is associated with disease. However, the dual nature of oxygen was recently demonstrated using a cell model with disruption of a gene critical for mitochondrial respiration
As the essential substrate for oxidative stress, ambient oxygen could play an important role in DNA damage and cellular senescence. Indeed, lowering the oxygen exposure of cultured cells has been shown to decrease oxidative DNA damage and extend cellular lifespan
Mice were bred and maintained in the approved animal housing facility of the National Institutes of Health, and all animal experiments were in accordance with the guidelines of the NIH animal care and use committee. The NIH ACUC approval identification number for this animal protocol is H-0126.
Antibodies used in the study are as follows: Rabbit polyclonal RAG1 (K-20, Santa Cruz Biotechnology); monoclonal phospho-S-139 histone H2AX (γ-H2AX, clone JBW301, Millipore); monoclonal β-actin (AC-15, Sigma); rabbit monoclonal β-catenin (6B3) and cyclin D1 (DCS6, Cell Signaling). Chemicals used in the study are as follows: avidin-FITC; 7,12-dimethylbenz[a]anthracene (DMBA); and phorbol 12-myristate 13-acetate (TPA) (all from Sigma).
The
A large hypoxia chamber (Coy Laboratory Product Inc.) capable of accommodating multiple standard-sized mouse cages with a live animal waste removal system was set at 10% O2, 0–0.5% CO2, 30–70% humidity at room temperature. These parameters were continuously monitored with internal and external probes through a tele-alarm service (Rees Scientific). All animal studies including endpoint determinations were performed in accordance with the guidelines and approval of the NIH animal care and use committee. In all animals used for tumor-free survival studies, necropsies were performed and histopathologic diagnoses confirmed by certified NIH veterinary pathologists.
C57BL/6J background strain wild-type and
Five week-old male
Five-wk old female FVB mice (FVB/NJ, Jackson Laboratory) were used for the two-stage chemical skin carcinogenesis model as previously described
Arterial blood oxygen saturation was determined using a pulse oximetry sensor (MouseOx™ Oxymeter, STARR Life Sciences Corp.) attached to the tail of 10% oxygen acclimated mice. Oxygen saturation was converted to partial pressure (mm Hg) as previously reported
Blood samples were collected in EDTA anti-coagulant tubes for hematocrit measurements. The Bruker Minispec NMR analyzer (Bruker Optics) was used to measure body composition of non-anesthetized mice after up to 16 wk of 10% oxygen exposure (∼21 wk old)
Athymic nude mice (B6.Cg-Foxn1nu/J, Jackson Laboratory) were subcutaneously injected with 10×106 cells into each hind limb and housed in the indicated oxygen condition. Tumor dimensions were measured using a digital caliper two times a week and volumes estimated as previously described (volume = (width2×length)/2)
Apoptotic cells were stained by the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) technique using paraffin-embedded tissue sections (Histoserve, Inc.). Images were captured using an Axioskop2 Plus fluorescence microscope with Axiovision software (version 4.6) and apoptotic nuclei were quantified.
Blood from wild-type C57BL/6J mice in room air or acclimated to 10% oxygen for at least 4 wk was collected in EDTA tubes and used for GSH and GSSG measurement according to the manufacturer's protocol (BIOXYTECH® GSH/GSSG-412 kit, OxiResearch). Absorbance signals were collected using the Victor3 plate reader (PerkinElmer).
Samples obtained from mice exposed to different oxygen concentrations were handled as rapidly as possible. The thymus tissue was gently dissociated into cell suspension (PBS) using sequential filtration through a nylon mesh (Gelman Laboratory) and a 40 µm cell strainer (BD Falcon) and further washed with PBS to remove tissue debris. The isolated thymocytes were counted, stained with 5 µM CM-H2DCFDA (Invitrogen) for 15 min at 37°C, and washed with PBS. DCF signal was collected and analyzed using FACSCalibur and CELLQusest™ software (version 3.3, BD Biosciences).
Frozen tissue sections (10 µm) were fixed in 2% paraformaldehyde PBS, permeabilized in 0.2% Triton X-100 PBS, RNA eliminated by RNase A (10 µg/ml) treatment for 1 hr at room temperature, and incubated in avidin-FITC (∼1∶200, Sigma) as previously described
Genomic DNA from tissues was isolated using the DNeasy kit (Qiagen) and 8-oxoG content determined using an ELISA kit according to manufacturer's protocol (8-OHdG Check Ultrasensitive ELISA, Multispecies Specificity, BioVendor). Absorbance signals were collected using the Victor3 plate reader (PerkinElmer).
Telomere length using 10 ng of purified genomic DNA was determined by a PCR-based technique as previously described
Telomere primers:
F:
R:
F:
R:
Tissues were homogenized in cold RIPA protein lysis buffer with protease inhibitor cocktail and frozen on dry ice for storage. Protein samples were heated for 5 min at 95°C prior to resolving and transferring in Novex Tris-Glycine gel and XCell II™ blot module (Invitrogen), respectively. Membranes were incubated overnight with primary antibody at 4°C and developed with HRP-conjugated secondary antibody (∼1∶10,000, Jackson Laboratory) using standard protocols.
mRNA levels of the HIF1-α target genes
VEGF primers
F:
R:
LDHA primers
F:
R:
To test the concept of ambient oxygen as a tumor promoter, we placed tumor-prone p53 deficient (
We first confirmed that tissue oxygenation is indeed decreased by hypoxia but that it is adequate to supply general metabolic needs. Notably, the partial pressure of oxygen in arterial blood was reduced by 50% as would be predicted from the magnitude of reduction in ambient oxygen (from 21% to 10%) while the hematocrit level was increased confirming chronic adaptation to hypoxia (
We next sought to provide an explanation for the improved tumor-free survival under the mildly low oxygen condition. It is formally possible that the decreased oxygen concentration in the hypoxia chamber could bioenergetically limit the growth of cancer cells by decreasing oxidative phosphorylation. To test this possibility, we examined the growth rate of the established human colon cancer HCT116 cells as xenografts in 10% versus 21% oxygen. The xenograft growth rates were similar indicating that the
During the course of our study, the majority of
Our current observation of increased tumor-free survival under reduced ambient oxygen in
Glutathione is the major free radical scavenger and an important marker of oxidative stress. To show that mice exposed to a lower ambient oxygen level had less oxidative stress, we measured both reduced (GSH) and oxidized (GSSG) glutathione in blood of mice chronically adapted to 10% oxygen. The GSH concentration increased by 43% in the 10% oxygen condition compared to room air while there was no significant difference in GSSG levels within the detection limits of the assay (
Oxidative stress has also been associated with telomere shortening
To further expand on our initial observation that oxygen promotes genomic instability and tumorigenesis, we examined the effect of ambient oxygen on the
As with the
Hypoxia causes physiologic adaptations to improve tissue oxygenation such as the observed increase in hematocrit. However, the oxygen in the superficial layer of the skin is largely supplied by direct exposure to air so that tissue oxygenation can be more predictably altered by changing ambient oxygen conditions
Under the 10% oxygen condition, there was a marked reduction in the formation of skin papillomas at both the standard (5 µg) and high (100 µg) dose of DMBA (
By reducing ambient oxygen exposure in three separate
We have previously shown that increased intracellular oxygen levels caused by the disruption of mitochondrial function may contribute to increased ROS generation and genomic DNA damage
Although our results show that lowering oxygen reduces genomic instability and tumorigenesis
Supraphysiologic levels of oxygen, including hyperbaric conditions, have been shown to inhibit tumorigenesis
The lessons from our current study may be applicable to human health. Supplemental oxygen is ubiquitously employed in clinical medicine because of its immediate benefits for energy production while the less apparent potential for genotoxicity can be neglected. Our work may provide a biological mechanism for important clinical observations such as the increased cancer risk of neonates exposed to supplemental oxygen or of babies conceived through
Blood hematocrit is increased after chronic adaptation to 10% oxygen. Data are shown as mean ± SEM, with n = 6 to 9.
(TIF)
No significant difference in body weight, body mass composition or food intake in the 10% versus 21% oxygen condition.
(TIF)
Lymphoma characteristics of
(TIF)
Expression levels of HIF1-α target genes are not significantly increased in 10% versus 21% oxygen. The relative levels of VEGF and LDHA mRNA as markers of HIF1-α activity were measured in liver, spleen and thymus of
(TIF)
We thank P. Amornphimoltham, L. Cao, M. Allen, J. Smith, and I. Rovira for advice and technical support. We are grateful to Toren Finkel for helpful suggestions and critical reading of this manuscript.