Making Hypoxia Happen

Photo: Courtesy of Patti Oprysko HYPOXIA IN ACTION: The ductus arteriosus of a newborn baboon showing EF5 binding (red), tissue perfused by blood flow (blue) and blood vessels (green). The ductus arteriosus is an artery, which bypasses the lungs before birth and must close to allow proper oxygenation of blood by the lungs after birth. Mount Everest climber Frank Smythe stood at 27,000 feet, near the top of the world, in 1933. Later, Smythe recounted an exceptionally odd experience. He saw "two curious looking objects floating in the sky. [One] possessed what appeared to be squat underdeveloped wings, and the other a protuberance suggestive of a beak. They hovered motionless but seemed slowly to pulsate."1 But in reality, he saw nothing; he just lacked oxygen. Into Thin Air's author Jon Krakauer's own Everest hypoxic delusion involved a green cardigan, wingtips, and an eerie bodily detachment.1 Most people will never experience such a dramatic lack of oxygen; even so, less than "normal" levels of oxygen can have profound physiological effects throughout the body. Just what is normal? Ambient air at sea level contains approximately 21% oxygen (O2) (~150 mmHg or torr), but the highest partial pressure (pO2) detected in the human body is around 16% (100 torr) in the lung alveoli.2 Throughout the rest of the body, pO2 values measure less than or equal to 6% (40 torr). In other words, "normoxia"--normal oxygen tension--is not constant, and neither, therefore, is hypoxia. The question has significant implications, as oxygen concentration affects many cellular processes and regulates many genes. This past January, Adrian L. Harris listed 90 hypoxia-inducible genes, in such diverse categories as oxygen transport, iron metabolism, angiogenesis, glycolysis, cell growth and inhibition, pH balance, neurotransmitters, cytokines, stress responses, cell adhesion, extracellular matrix, cytoskeleton, proteases, and coagulation.3 Hypoxia is clearly at work in areas beyond the intuitive ones of ischemic, vascular, and lung disease. Though oxygen-regulated gene expression receives most of the limelight, one issue rarely addressed is, how do researchers create and measure hypoxia in the lab? INVASION OF THE OXYGEN PROBES Hypoxia research is most advanced relative to cancer research, owing to the pivotal role low oxygen tension plays in the development of radiation resistance and tumor regrowth. In their seminal demonstration of chronic hypoxia more than 50 years ago, L.H. Thomlinson and R.H. Gray examined hundreds of tumor sections histologically and observed a consistent, gradual increase in necrotic lung tumor cells as the distance from blood vessels increased. They concluded that the limiting factor for cell viability was the distance required for oxygen and nutrient diffusion.4 In Radiation Biology for the Radiologist, Eric Hall describes the "oxygen fixation hypothesis" wherein O2 is required not only to produce the radical oxygen species that damage DNA and result in cell killing, but also to prevent repair of that damage. Thus, hypoxic tumor cells have a distinct survival advantage; they can repair the minimal DNA damage created by radiation therapy in a low-oxygen microenvironment, and it takes two to three times the dose to kill them, which the surrounding normal tissues cannot tolerate.5 Hypoxic cells in tumors contribute to radiation and chemotherapy resistance and subsequent recurrence to such an extent that human clinical trials are underway in the United States, Canada, the United Kingdom, and Europe. The therapeutic ramifications of screening for tumor hypoxia are significant. "Once you identify the patients [who] have hypoxic tumors, then you have the population for testing potential therapies," says Sydney Evans, associate professor of radiation oncology at the University of Pennsylvania Medical School and coordinator of ongoing clinical trials involving head and neck squamous cell carcinoma, soft-tissue sarcomas, cervical tumors, and brain tumors. The first trials evaluating the correlation of hypoxia to therapeutic outcome were made possible by the development of the Eppendorf pO2 Histograph.6 This commercially available technology, recently licensed to Helzel Medical Systems of Kaltenkirchen, Germany, is a mobile unit carrying a computer-driven polarographic needle electrode equally at home in an operating suite or a research environment. According to North American distributor, Mary-Claire Kavanagh, Helzel has upgraded the system by providing a better user interface for the rechristened Phoenix THM (tissue hypoximeter). Once positioned and programmed, the Phoenix probe advances, retreats, then measures pO2 in a continuous stepping path through the tissue, outputting an oxygen histogram. Researchers can also probe oxygen tension using immunohistochemistry. The US Food and Drug Administration recently approved use of the nitroimidazoles, pimonidazole, and EF5, in clinical trials. According to James Raleigh, professor of radiation oncology, University of North Carolina Medical School, "The initial study [with pimonidazole] was to do a survey of hypoxia in different human tumors and to demonstrate the presence, frequency, and extent of hypoxia." Studies of hypoxia-induced gene expression and correlations with clinical outcome are in progress. Patients receive EF5 or pimonidazole prior to surgery or biopsy, allowing time for distribution, selective activation through bioreduction, and subsequent binding in hypoxic regions.7 Researchers can then identify the hypoxic regions and correlate their locations with other biomarkers, such as vascular endothelial growth factor or hypoxia-inducible factor-1 (HIF-1), using immunohistochemical methods. The nitroimidazole immunohistochemical assays for hypoxia are powerful tools outside the clinical arena, and scientists can order the reagents and antibodies online (EF5 reagents and protocols: hypoxia.xrt.upenn.edu; pimonidazole [Hydroxyprobe-1™] kits and background information: www.radonc.unc.edu/pimo). NONINVASIVE TECHNIQUES Tissue section assays and oxygen probe measurements have potential use as prognostic or predictive biomarkers for more individualized cancer treatment; however, both are invasive methodologies. Tumor hypoxia measurement is limited to surgical cases and sites accessible to biopsy. Maureen Thorniley, senior lecturer at the University of Manchester Institute of Science and Technology, and this year's president and host for the annual meeting of the International Society of Oxygen Transport in Tissue, reports that the most active area of interest is the development of completely noninvasive oxygen-dependent imaging tools; but, some noninvasive and minimally invasive techniques already exist. Photo: Courtesy of James Raleigh PHOTOMICROGRAPH of a E106 human gliomablastoma multiforme xenograft growing in the brain of a nude mouse. Shown are vasculature (red), hypoxia (green) and perfusion (blue) in normal brain (N) and tumor tissue (T). (Reprinted with permission, Bernsen et al., Journal of Neurosurgery 93:449-54, 2000.) Scientists can use radiolabeled hypoxia markers to identify low-oxygen environments in tumors with positron emission tomography (PET), single photon emission computed tomography (SPECT), and magnetic resonance imaging (MRI).8 The results of these studies show promise, but imaging hypoxia using these minimally invasive techniques is complicated by metabolism of the parent compound, the actual binding process, the lifetime of the radionuclide, the partition coefficient of the drug, and drug clearance. Improvements in the hardware and software of PET and SPECT scanners should improve the resolution of hypoxia in tumors in the future.8 Both f-MRI BOLD (functional MRI blood oxygen level dependent) and NIRS (near-infrared spectroscopy) measure oxygen-dependent signals produced by the body's own chemistry. f-MRI BOLD detects magnetic properties of deoxy- hemoglobin, recording variations in blood oxygen flow and tissue oxygen consumption. NIRS measures the attenuation of light passing through tissue using an optical method to re-cord oxygen-dependent signals from hemoglobin, deoxyhemoglobin, and cytochrome-c oxidase. The NIRS system is portable,provides real-time data, and, according to Thorniley, "can be applied from cuvette up to man." Japanese companies Hamamatsu Photonics and Shimadzu Medical Systems sell these devices. Although each of these technologies can provide valuable hypoxic measurements, the data they yield--with the exception of NIRS--are captured as a snapshot. Real-time or continuous pO2 readout would enhance researchers' understanding of the dynamics and heterogeneity of oxygen tension in normal, tumor, or other diseased tissue physiology. The first of these real-time assays for pO2 is phosphorescence lifetime imaging based on oxygen-dependent quenching of phosphorescence.7 After injection of a near-infrared-absorbing Pd-porphyrin probe, David Wilson and colleagues in the Department of Biochemistry and Biophysics at the University of Pennsylvania Medical School, excite phosphorescence with light through a light guide. The more oxygen present in the microenvironment of the phosphor, the faster the signal decays, emitting less light. As pO2 declines, the phosphorescence lifetime and intensity increases. This results in a higher signal-to-noise ratio with greater sensitivity at lower pO2 values. Phosphorescence measurements can be taken at the surface or through tissue and can be made as quickly as several per second, or at longer intervals, as long as the phosphor is present. "Because it's an optical method, in thin surface tissue like the retina of the eye," says Wilson, "you can make very high-resolution images of the distribution of oxygen within that tissue in two dimensions." This kind of dynamic oxygen mapping in the retina is being used to study diabetic retinopathy and age-related macular degeneration. Wilson concludes, "There is an enormous amount of pathology in virtually every tissue you can imagine which may well be hypoxia-related; but, in the absence of measurements, often hypoxia is not considered as a cause." The Phosphorescence O2 Measurement System requires calibration only once to deliver absolute measurements, providing experimental conditions remain constant. The equipment and reagents are produced by Philadelphia-based Oxygen Enterprises; some are available as well through Holliston, Mass.-based Harvard Apparatus, which also carries a varied line of other chambers and tools for oxygen research. Oxford, UK-based Oxford Optronix, markets a fiberoptic fluorescence lifetime probe system, OxyLite™ and OxyLab pO2™, designed primarily for in vivo use. According to Michael Rau, production and applications specialist, a rapidly pulsing light-emitting diode (LED) illuminates a fluorescent dye containing ruthenium that fills the biosensor's tip. As in the phosphorescence assay, the fluorescence lifetime is inversely proportional to the concentration of oxygen. Thus, both assays offer greatest sensitivity at lower pO2. The OxyLite is precalibrated, MRI compatible, and available with a probe-integrated thermocouple to monitor pO2 in hyperthermia experiments. Depending on the laser used, the device can monitor oxygen tension in the relatively new field of photodynamic therapy research. Oxford-Optronix's Web site is particularly useful and informative; it provides sample data in various model systems, many references, and detailed explanations. Ocean Optics in Dunedin, Fla., offers a similar fluorescence probe called the FOXY Fiber Optic Oxygen Sensor. According to Mahmoud Shahriari, Ocean Optics' director of the sensors division, optical probes are especially useful "since they are immune to electromagnetic wave interference." CREATING HYPOXIA Measuring hypoxia accurately is only half the story. Creating reproducible hypoxia in vitro or in vivo can be technically challenging. "There are just so many physical factors involved in how a dish comes into equilibrium with its gas phase, such as diffusion, convection, and mixing," explains Cameron Koch, professor of radiation oncology at the University of Pennsylvania School of Medicine and EF5's co-inventor. Other general complicating factors include control of cellular and chemical oxygen consumption.9 In the early 1970's, spearheaded by work in England by J. Donald Chapman and J.W. Boag, "researchers showed the properties of all the different plastics that would lead to large amounts of oxygen coming out of them at very slow rates," continues Koch. A serendipitous suggestion by a physicist at a meeting inspired Koch to build his own "hypoxia chamber" out of aluminum instead of plastic. He developed a thin-layer film technique9 using inert glass dishes and slow shaking to speed convection, combined with optimal cell numbers and buffers to create a hypoxic environment based on Boag's mathematical model. Undaunted, intrepid hypoxia research-ers use several different kinds of equipment to work with cells, tissues, and whole animals. The first stop for Sean Colgan, associate professor of anesthesiology, Harvard Medical School, was the scientific literature. There, he identified a colleague working successfully with hypoxia who suggested using the vinyl hypoxia glove box produced by Coy Laboratory Products of Grass Lake, Mich. Colgan's current Coy model is outfitted with a temperature controller, a humidity pan, a fish tank pump to circulate hypoxic air, and a flow meter. For short-term hypoxia, researchers often anesthetize animals and attach a nose cone carrying the hypoxic blend of gases. But, Colgan notes, the commonly used anesthetic ketamine is also an anti-inflammatory agent; anesthesia itself can, therefore, actually affect hypoxia experiments. Colgan and many others use San Diego-based Billups-Rothenberg's MIC-101 modular incubator chamber to hold tissue culture vessels or mice in hypoxic conditions. A recent collaboration between Coy Laboratory Products and Billups resulted in a glove box with an air-locked entry large enough to transfer the MIC-101 incubator into Coy's hypoxic glove box. This permits additions or sample withdrawals without loss of hypoxic pO2. Coy is in the process of developing a mini-incubator within its glove box. BioSpherix (formerly Reming Bioinstruments) of Redfield, NY, offers two products for hypoxia research. The PRO-OX 110 regulates oxygen tension in either tissue culture incubators or animal containment. For more complex studies of chronic, acute or intermittent hypoxia, the company offers the OxyCycler®. Winchester, Va.-based Jouan markets a three-gas incubator, the IG750, that permits researchers to sustain specific O2 set points. However, maintenance of O2 tensions of 1% or lower requires large amounts of nitrogen and door openings will perturb the hypoxic environment. A decidedly lower-tech approach is chemical hypoxia, caused by interference with HIF-1degradation. Addition of desferrioxamine or cobalt chloride to tissue culture medium activates HIF-1-regulated signal transduction. This method is usually used alongside a gas approach, but may not produce exactly the same results. Perhaps the most pleasant way to experiment with hypoxia is to visit North America's highest research facility. The University of California's White Mountain Research Station rents lab space at 14,249 feet. Call for rates and availability. Susan Jenkins can be contacted at sjenkins@the-scientist.com. References 1. J. Krakauer, Into Thin Air--A personal account of the Mt. Everest disaster, New York: Anchor Books, 1997, pp. 251-2. 2. G.L. Semenza, "Hypoxia-inducible factor 1: Oxygen homeostasis and disease pathophysiology," Trends in Molecular Medicine, 7:345-50, August 2001. 3. A.L. Harris, "Hypoxia--a key regulatory factor in tumour growth," Nature Reviews Cancer, 2:38-47, January 2002. 4. R. Thomlinson, L. Gray, "The histological structure of some human lung cancers and the possible implications for radiotherapy," British Journal of Cancer, 9:539-49, 1955. 5. E. Hall, Radiobiology for the Radiologist, 5th ed., Philadelphia: Lippincott, Williams & Wilkins, 2000, pp. 96-9. 6. M. Hockel et al., "Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix," Cancer Research, 56:4509-15, 1996. 7. M. W. Dewhirst et al., "Review of methods used to study oxygen transport at the microcirculatory level," International Journal of Cancer (Radiation Oncology Investigations), 90:237-55, 2000. 8. J.R. Ballinger, "Imaging hypoxia in tumors," Seminars in Nuclear Medicine, 31:321-9, October 2001. 9. C.J. Koch, "Measurement of absolute oxygen levels in cells and tissues using oxygen sensors and the 2-nitroimidazole EF5," Methods in Enzymology, 352:3-31, 2002.

Making Hypoxia Happen

Interested in reading more?

Become a Member of