| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Commentary |
Department of Biomedical Physics and Bioengineering, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK
Mention medical imaging and most radiologists and medical physicists will think of CT, MRI, ultrasound and nuclear medicine, the techniques that form the backbone of diagnostic radiology. Nevertheless, there are a number of other methods currently under development that are potentially capable of producing extremely valuable functional information. One of these up-and-coming areas is free radical imaging. Although the techniques used are based on MRI, free radical imaging is sufficiently different that many MR physicists and radiologists may be unaware of the methodology and of its potential uses. This commentary describes the techniques currently in use or under development and will highlight some of the biomedical applications for which they have been used and for which they may be used in the future.
Free radicals are molecules with one or more unpaired electron in their outer orbitals. A number of related MR methods are used for imaging free radicals, all of which make use of the fact that a free radical's unpaired electron exhibits a quantum-mechanical spin and therefore has a magnetic moment. This means that an MR signal can be generated in a manner exactly analogous to the detection of hydrogen nuclei (protons) by nuclear magnetic resonance (NMR) in conventional MRI. MR of unpaired electrons is called electron spin resonance (ESR). Substances that have unpaired electrons are termed paramagnetic, thus ESR is also referred to as electron paramagnetic resonance (EPR). The mass of an electron is about three orders of magnitude smaller than that of a proton, therefore the MR properties of the two particles are rather different. In a given strength of applied magnetic field, the ESR frequency is 659 times that of proton NMR. For example, in a magnetic field of 1 T (a common field strength for clinical MRI), the NMR frequency is 42.6 MHz while the ESR frequency would be 28 GHz, well into the microwave part of the electromagnetic spectrum, with strong absorption by conducting samples such as tissue. Owing to this, most biomedical free radical imaging by ESR is carried out using much lower magnetic field strengths, usually between 10 mT and 40 mT, with resonant frequencies between 250 MHz and 1 GHz, which can penetrate a few centimetres into tissues allowing at least small animals to be studied.
Another difference between ESR and NMR is that the electron relaxation times of free radicals are very short, typically between 0.1 µs and 1 µs, a million times shorter than those encountered in clinical MRI. As a result, most ESR spectroscopy and imaging is done using "continuous wave" (CW) detection methods. Instead of applying a pulse of radiowave energy and waiting for the transient response, as in clinical MRI, in CW ESR the sample is continuously irradiated with low intensity electromagnetic radiation and the resonant response of the unpaired electrons is measured by slowly increasing the strength of the applied magnetic field. When an ESR is encountered, the combined effect of the unpaired electrons (the "electron magnetization") alters the electrical properties of the resonator used to apply the radiowaves and a reflected signal can be measured. To obtain spatial information about the sample, a magnetic field gradient is applied continuously during the magnetic field sweep. An image is built up by back-projecting a series of one-dimensional projections of the sample, obtained by measuring the ESR signal repeatedly, with the direction of the applied magnetic field gradient stepped in small increments [1].
Perhaps the greatest technical difficulty of the CW ESR method is its extreme sensitivity to physiological motion of the animal under study, which results in both increased noise and image artefacts. One solution is to use sophisticated electronics to correct for changes brought about by animal motion, but in fact the problem can be almost completely alleviated by using an alternative detection scheme called longitudinally-detected ESR (LODESR). In this method, the sample is irradiated with electromagnetic radiation close to the desired resonant frequency (say 300 MHz), the intensity of which is modulated at a much lower frequency (say 0.5 MHz). This causes the unpaired electron magnetization to oscillate at twice the modulation frequency, and the signal is detected by a receiver coil oriented along the same direction as the applied magnetic field [2].
In both ESR and LODESR imaging the gradient strength must be sufficient to overcome the rather broad ESR lines, and the gradients used are generally at least an order of magnitude larger than those used in MRI. This, together with the reduced penetration depth at the resonant frequencies used, has so far limited whole-body imaging of free radicals to studies in mice or rats. In fact, it is extremely unlikely that ESR imaging will ever be possible on a whole-body human scale, although localized imaging may be possible and some research groups are already investigating high resolution skin imaging by ESR.
Another method of imaging free radicals is available that does not suffer from sample size limitation and, moreover, that can generate images whose resolution is independent of the ESR line width of the free radical under study. This technique is known as proton–electron double resonance imaging (PEDRI) [3]. The technique is based on the Overhauser effect, thus it is also known as Overhauser MRI. PEDRI uses a combination of ESR and MRI: the ESR of a free radical of interest is irradiated during the collection of an MR image. The Overhauser effect causes an increase in the NMR signal strength in parts of the sample containing free radicals, and these regions are revealed by an increased intensity in the final image. PEDRI uses standard MRI software and hardware, but requires the additional capability of irradiating the sample at the ESR frequency. The basic PEDRI technique still suffers from the need for the ESR irradiation to penetrate the sample, so it is implemented at low field and is limited to small animals in the same way as ESR or LODESR imaging. However, field-cycled PEDRI (FC-PEDRI) overcomes this limitation, and could potentially allow free radical imaging in humans. It works by switching the applied magnetic field between two values during the MRI pulse sequence. A very low magnetic field (say 3 mT) is applied while ESR irradiation is taking place, so that this can be at a low frequency (
50 MHz) that can easily penetrate the body. The field is then increased to a much a higher value (say 60 mT) so that the NMR signals can be measured with improved signal-to-noise ratio and therefore good image quality [3].
Two of the most important applications of free radical imaging are endogenous free radical imaging and oxygen concentration imaging using exogenous free radical "probes". Important developments have been made in both of these areas during the last 5 years. Naturally occurring or endogenous free radicals are vital to our survival, as many of them are involved in normal metabolism. Thus, free radicals such as hydroxyl and nitric oxide are always present in the body, although normally at a very low concentration. There is a large body of evidence that the onset of many diseases is accompanied by a significant change, often a rise, in the concentration of some endogenous free radicals. The list of disease states thought to involve free radicals is extensive and includes inflammatory conditions, ischaemic disease and possibly cancer. However, the evidence for free radical involvement is mainly indirect, being obtained from biochemical analysis of body fluids or tissue samples, for example. Free radical imaging offers the possibility of detecting and identifying free radicals in situ, with obvious benefits for biomedical research and, possibly, for diagnosis.
As the sensitivity of free radical imaging technology has increased, it has become possible to image the production of nitric oxide in the bodies of animals subjected to various types of stress. The first demonstration of this exciting application was published in 1996 by a research group from Yamagata, Japan. Mice were given a dose of lipopolysaccharide and the resulting nitric oxide produced in the liver was imaged by ESR [4]. More recently, the same group imaged the generation of nitric oxide in mice given a dose of a nitrovasodilator compound [5]. In both cases a method called spin-trapping was used to stabilize the short-lived ·NO molecules, enabling a sufficient concentration to build up for the imaging experiment to succeed. The use of spin-trapping isessential in most studies of short-lived endogenous free radicals. However, it is not always necessary to introduce chemicals to achieve this effect, as in some cases naturally occurring substances can form long-lived complexes with endogenous free radicals, which can then be detected by MR [6].
Oximetry by ESR was proposed more than 25 years ago, but its practical implementation as an imaging method has had to await technical developments that have only recently been realized. The technique relies upon the fact that the presence of dissolved molecular oxygen has a strong, usually reversible, influence on the ESR relaxation times of paramagnetic "probe" materials. Thus if images can be collected that show the spatial distribution of the ESR line width, it is possible to generate oxygen concentration images from these data. Recent results have demonstrated the ability to perform these experiments in the tail veins and arteries of living rats [7], and there is no doubt that the same methodology will be applicable in the whole animal before long. An extensive review of oximetric techniques has recently been published [8].
Most in vivo free radical imaging to date has focused on the use of exogenous free radicals, especially nitroxides. These compounds are stable in solution and can have low toxicity and relatively long biological half-lives (30 min), depending on their chemical structure. As well as providing oximetric information, free radical imaging with nitroxides can provide functional information by, for example, measuring the clearance rate of a nitroxide through an animal's kidneys. Nitroxide chemistry is extremely versatile, and work is going on to develop pH sensitive nitroxides that will allow local pH levels to be monitored in the body by measuring the spatial variation of the compounds' ESR spectra. Some of these pH sensitive nitroxides have already been examined in vivo, with promising results [9].
Compared with MRI, the research field of free radical imaging is still very small, with perhaps 20research groups worldwide working on the development of ESR and LODESR imaging and PEDRI. Nevertheless, significant progress continues to be made in these fields, and it is encouraging that the number of research groups using free radical imaging in biomedicine is starting to grow. Developments to watch in the coming years include pulsed ESR imaging, which offers significantly reduced image acquisition times [10], and FC-PEDRI, which, with higher detection magnetic fields, offers improved sensitivity to allow lower concentrations of free radicals to be imaged. Free radical imaging is gaining acceptance as a valuable tool in biomedical research and may even find clinical applications in the future.
Received for publication November 29, 2000. Accepted for publication May 25, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  K. Jiang, C. Schwarzer, E. Lally, S. Zhang, S. Ruzin, T. Machen, S. J. Remington, and L. Feldman Expression and Characterization of a Redox-Sensing Green Fluorescent Protein (Reduction-Oxidation-Sensitive Green Fluorescent Protein) in Arabidopsis Plant Physiology, June 1, 2006; 141(2): 397 - 403. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
