Online Issues
<< All Back-issues
<< This Issue's Table of Contents
ILAR Journal V42(3) 2001
Impact of Noninvasive Technology on Animal Research
Positron
Use of Positron Emission Tomography in Animal Research
Simon R. Cherry and Sanjiv S. Gambhir
| Simon R. Cherry, Ph.D., and S. S. Gambhir, M.D., Ph.D., are Associate Directors of the Crump Institute for Molecular Imaging and Associate Professors in the Department of Molecular and Medical Pharmacology in the University of California Los Angeles School of Medicine. |
Among the several imaging technologies applied to in vivo studies of research animals, positron emission tomography (PET) is a nuclear imaging technique that permits the spatial and temporal distribution of compounds labeled with a positron-emitting radionuclide to be determined noninvasively. It can be viewed as an in vivo analog of classic autoradiographic methods. Many different positron-labeled compounds have been synthesized as tracers that target a range of specific markers or pathways. These tracers permit the measurement of quantities of biological interest ranging from glucose metabolism to gene expression. PET has been extensively used in imaging studies of larger research animals such as dogs and nonhuman primates. Now, using newly developed high-resolution dedicated animal PET scanners, these types of studies can be performed in small laboratory animals such as mice and rats. The entire whole-body biodistribution kinetics can be determined in a single imaging study in a single animal. This technique should enable statistically significant biodistribution data to be obtained from a handful of animals, compared with the tens or hundreds of animals that might be required for a similar study by autoradiography. PET also enables repeat studies in a single subject, facilitating longitudinal study designs and permitting each animal to serve as its own control in experiments designed to evaluate the effects of a particular interventional strategy. This paper provides a basic overview of the methodology of PET imaging, a discussion of the advantages and drawbacks of PET as a tool in animal research, a description of the latest generation of dedicated animal PET scanners, and a review of a few of the many applications of PET in animal research to date.
Key Words: animal models; imaging; microPET; PET; positron emission tomography
Introduction
Animal models of human disease are an important and widely used research tool in understanding disease processes and evaluating potential therapies. Imaging technologies are playing a growing role in animal research, enabling large, expensive laboratory animals to be studied noninvasively and providing the possibility of reducing the number of smaller laboratory animals required in typical longitudinal studies. As each animal serves as its own individual control, the reproducibility of data from imaging studies may actually be better than that obtained from traditional invasive techniques, although this improvement has yet to be demonstrated unequivocally. Finally, imaging provides a bridge from animal research to human research and the clinic, enabling similar and sometimes identical experiments to be carried out across species. This advantage could be very important in understanding the value and limitations of animal models and in further refining these models and their extrapolation to the human.
Genetically modified animals have become a particular focus for advancing our understanding of mammalian biology. The mouse is the experimental model of choice in many situations, due to a range of factors including relatively high genetic homology with man and widespread expertise and well-developed methodology for genetic manipulation. Furthermore, the ability to use nude or severe combined immunodeficiency disorder mice with xenografts of human tissue make the immunodeficient mouse model particularly attractive. Compared with other mammalian models, the fast breeding cycle and decreased housing and maintenance costs are additional attractions. In some disciplines, particularly in neuroscience, the rat also remains an important experimental animal. Its value is due to the availability of well-established experimental models in the rat that predate much of mouse genomics and the larger brain, which allows better stereotactic accuracy in models involving surgical or other invasive procedures directed toward specific brain regions.
Much of the technology development in imaging is now directed at these small laboratory animals. However, larger laboratory animals such as dogs, pigs, and nonhuman primates will continue to play an important role in select experiments, particularly those involved in studying the central nervous system and the heart. In all cases, whether in a mouse or a primate, imaging offers opportunities to reduce the number of animals needed to answer a specific question, to refine experimental protocols by allowing longitudinal within-subject design, and possibly to lower the cost and reduce the overall time needed to obtain a result. Furthermore, for the first time, experimental studies in which the hypotheses being studied can be answered only by noninvasive experimentation are now possible (e.g., studying the development of new blood vessels to allow tumor growth).
Positron emission tomography (PET1) is an imaging technique that employs radioactively labeled compounds to study a wide range of biological processes in vivo (Phelps 2000). In this article, we explain the essential properties of PET and compare them with other imaging techniques that use some form of targeted contrast agent to measure biological function. We describe PET imaging systems that have been developed specifically for animal research, and we provide an overview of some of the applications of PET in animal research.
Imaging Probes and Contrast Agents
Labeling of relevant molecules so that their distribution can be visualized in biological systems is a widely used method in the laboratory. The most commonly used labels are fluorophores, which emit light when stimulated with light of the appropriate wavelength, and radionuclides, which emit gamma rays or beta particles when they decay. In cell culture and in tissue slices (e.g., in situ hybridization), both radioactive and fluorescent probes are used very successfully to study a whole range of biological problems. However, these model systems are highly simplified. The ability to track these "tagged' molecules in vivo is valuable because it enables the animal to be studied at the level of the whole organism, including the effects of transport, metabolism, excretion, and interactions and signaling between organ systems. This broader context is important both in studying normal animals under a variety of conditions or with different stimuli and in gaining a more detailed understanding of biological processes in animal models of disease. The factors mentioned above often are determinants of therapeutic efficacy in the treatment of disease, so their inclusion in the animal model is clearly advantageous as new therapies are being developed and tested.
Radiolabeled molecules can be examined in vivo by administering them by injection into a living animal; however, the distribution of these probes is commonly determined by euthanizing the animal followed by tissue sectioning and exposure to film or digital imaging plates (e.g., autoradiography) or direct tissue counting in a radiation detector. Each animal can therefore provide only a single snapshot rather than a continuous movie of the tracer distribution after injection. Furthermore, studies that examine the effect of some kind of perturbation on the system (e.g., pharmacological intervention, surgical manipulation, exposure to environmental factors) must use different animals for the control and experimental groups. A complete pharmacokinetic or pharmacodynamic time course, after interventional effects on an animal model over a period of several weeks, can easily require the use of several hundred animals.
It is clear that noninvasive techniques that provide the same or similar information without the need to euthanize the animal have great value. Imaging provides a pathway to do this (Weissleder 1999), and techniques that utilize tagged molecules such as optical and nuclear imaging and that can be used to investigate a wide range of biological processes are particularly powerful. Other techniques such as magnetic resonance imaging and x-ray computed tomography provide exquisite high-resolution images that are largely reflective of anatomy. These techniques also have an important place in small animal imaging but are in general more optimized to address different sets of questions. Some limited functional information can be obtained with modern magnetic resonance imaging methods, although the sensitivity levels are typically in the milli- to micromolar range, whereas the nuclear and optical methods can provide sensitivities beyond the nanomolar range. Sensitivity in this context refers to the number of tagged molecules that must be present in a given volume element to be detectable relative to a "low" background.
A number of groups are developing optically based techniques for in vivo imaging. One approach images fluorescent probes by injection into an animal (Weissleder et al. 1999). A high-sensitivity charged-coupled device (CCD1) camera is used to record the light emitted when these probes are stimulated by an external light source. By using fluorophores in the near infrared, the emitted light can penetrate several millimeters of tissue, and it has been possible to obtain projection (but not tomographic) images of the distribution of fluorescent compounds in vivo. Other approaches in genetically modified animals include the use of genes that encode for fluorescent proteins (e.g., green fluorescent protein) or for enzymes responsible for bioluminescence (e.g., the luciferase enzyme) (Contag et al. 1998). Both of these approaches result in optically detectable signals from cells in which the gene is being expressed. Although these optical techniques have the advantage of relative low cost without involving radioactive materials, they as yet provide largely qualitative data in vivo, unless the emission is located very close to the surface. This is a consequence of the high scattering and absorption of light in even thin sections of tissue, the highly depth-dependent distribution of light escaping from the animal's body, and the lack of tomographic methods for computing the three-dimensional emission distribution accurately. These factors also limit in vivo optical imaging methods to the smaller laboratory animals (mice) or to superficial imaging in larger animal models.
In vivo nuclear imaging involves labeling compounds with radionuclides that decay by the emission of one or more gamma rays. These radionuclides have the advantage that the penetrating gamma radiation can easily escape from the tissues of both small and large animals to be registered by an external detection system. Thus, it can be used in all experimental animal models and can penetrate through many centimeters of tissue. Furthermore, by taking multiple views around the animal, it is possible to use classical computed tomography methods to reconstruct quantitative three-dimensional images of the distribution of the compound of interest. Nuclear imaging techniques have the disadvantage of being somewhat more expensive and lacking control over the emission of the radiation. In contrast, in fluorescent imaging, light emission is stimulated by a laser or other external light source, and molecules labeled with multiple fluorophores can be converted from a quenched state (where no light is emitted) to an emitting state by specific biological events (Weissleder et al. 1999). From an imaging perspective, one distinct advantage of an optical approach is that in many applications, the background signal can be minimized because the optical signal can be nonexistent or quenched until the optical probe comes into contact with some target tissue. With radioactive approaches, it is not possible to control when the probe is radioactive or not; only the passage of time can allow background signal to clear.
Two forms of nuclear imaging can be used for in vivo small animal imaging. One uses single radionuclides that emit single gamma rays such as 99mTc, 123I, 201Tl, and 111In. The energy of emissions is typically in the 80- to 350-keV range, and these gamma rays can be detected externally by position-sensitive scintillation detectors, also known as gamma cameras. The gamma camera also contains a collimator, which is essentially a lead disk containing a large number of small parallel holes that allow only gamma rays arriving perpendicular to the camera face to be detected. All other gamma rays are absorbed by the lead walls between the holes. Although it is an inefficient use of the emitted radiation, the collimator is necessary to define the direction from which the gamma rays are coming, thereby providing spatial information. These gamma camera systems produce two-dimensional projection images of the distribution of a radionuclide. If the gamma camera is rotated about the object and sufficient views are collected, three-dimensional tomographic images can also be reconstructed. This technique is known as single photon emission computed tomography imaging and is discussed in the context of small animal imaging in a review by Weber and Ivanovic (1999). The other form of nuclear imaging, which uses positron-emitting radionuclides, is discussed in detail below.
Positron Emission Tomography: Basic Features
PET relies on the unique decay characteristics of positron-emitting radionuclides (Jones 1996; Phelps 2000). When a positron (the antiparticle to the electron with the same mass but with opposite electric charge) is emitted from the nucleus, it travels a short distance (typically a few tenths of a millimeter) before undergoing an annihilation reaction with an electron in tissue. The annihilation of the two particles results in the simultaneous emission of back to back gamma rays, as shown in Figure 1. These gamma rays carry energy of 511 keV, independent of the radionuclide involved. A list of some commonly used positron-emitting radionuclides and their characteristics appears in Table 1 along with other longer half-life positron-emitting radionuclides that may find increasing application in animal research.
If the two simultaneously released gamma rays are registered by external detectors placed around the subject (Figure 1), then each detected gamma ray pair defines a line (with a small uncertainty due to the travel of the positron) through the imaging volume along which the decay of the nucleus must have occurred. Because two gamma-rays with a strict geometric relation are emitted, imaging of positron-emitting radionuclides does not require a collimator. In a PET imaging study, large numbers of gamma-ray pair events are collected (typically 106-107) and reconstructed into cross-sectional images using standard computed tomography reconstruction methods such as filtered backprojection. More sophisticated statistical algorithms are also available to reconstruct the data (Ollinger and Fessler 1997). The resulting images, assuming that appropriate corrections are applied to the data, quantitatively reflect the distribution of the positron-emitting radionuclide in the body.
The detectors in most PET scanners, both animal and human, are based on some form of position-sensitive scintillation detector, using either small individual crystals of bismuth germanate scintillator or a continuous sheet of sodium iodide scintillator connected to photomultiplier tubes. The probability level of the 511 keV gamma rays interacting in these dense, optically transparent scintillator materials is somewhat high, and some of the energy deposited by the gamma rays is released as a flash of visible light. This light is then collected by a photomultiplier tube connected to the scintillator, which converts the light photons into electrons and amplifies the signal by a factor of approximately 106. Thus, the scintillation detector converts the interaction of a 511 keV gamma ray in the scintillator into a robust current pulse that can be detected and processed by relatively standard electronics.
The spatial resolution of a PET scanner is determined by the size of the scintillator crystals or, in the case of continuous sheets of scintillator, the thickness of the sheet and the size of the photomultiplier tubes reading out the scintillator sheet. Data collected from a PET scanner is not directly in the form of a tomographic image but instead, must first be reconstructed using the well-established methods of computed tomography (the same reconstruction mathematics is used by computed tomography scanners). The signal-to-noise level in the reconstructed PET images depends on the number of gamma ray events that were detected and formed the data set. The more events that are collected, the better the statistical quality of the data and the lower the noise level in the images. The signal-to-noise level is therefore determined by the solid angle coverage of the PET scanner (the gamma ray pairs are emitted isotropically), the efficiency of the detectors for stopping the gamma rays, the injected dose, and the imaging time.
One of the key advantages of PET is the availability of positron-emitting isotopes of carbon (11C; T1/2 = 20 min), nitrogen (13N; T1/2 = 9.9 min), oxygen (15O; T1/2 = 2.1 min), and fluorine (18F; T1/2 = 110 min). One or more of these elements are found in most biologically active compounds. Their presence allows radiotracers to be created by direct isotopic substitution (e.g., a positron-emitting C-11 atom substituted for a stable C-12 atom). In this manner, radiolabeled versions of almost any molecule of biological significance can be synthesized without altering the biological function of the molecule. Sometimes the structure of biologically relevant molecules are deliberately altered to create labeled analog tracers that have superior properties for the purposes of imaging in comparison with the native molecule. For example, an analog may isolate particular metabolic steps or may show improved uptake or clearance properties. One of the challenges in using the radionuclides mentioned above is their short half-life. This fast decay requires that radiolabeled molecules are synthesized very rapidly and that imaging takes place within one or two half-lives of the completion of the synthesis.
A second challenge related to the short half-lives of many important positron-emitting radionuclides is that it is usually necessary to produce them on site in a compact biomedical cyclotron (McCarthy and Welch 1998). 18F has a sufficiently long half-life that it can be distributed from regional distribution centers where a single cyclotron serves multiple customers within a particular area. There are also several longer-lived radionuclides (e.g., 64Cu, T1/2 =12.7 hr; 124I, T1/2 = 4.2 days) that can be shipped from remote sites and are particularly well suited for studying biological processes in which tracer accumulation and/or clearance of nonspecifically bound tracer requires many hours. A third option are generator-produced radionuclides (e.g., 62Cu, T1/2 = 9.7 min; 68Ga, T1/2 = 68 min), which consist of a long-lived parent radionuclide (produced at a remote accelerator or reactor) that continuously decays into a shorter-lived positron-emitting daughter radionuclide (Knapp and Mirzadeh 1994). The generator consists of a column containing the parent radionuclide, which can be flushed to elute the shorter-lived positron-emitting radionuclide whenever it is required. The use of either long-lived or generator-produced radionuclides enables PET studies to be carried out without an on-site cyclotron, although the radionuclides available are generally difficult to incorporate into molecules that are biologically relevant, which restricts the range of studies that can be performed.
A wide array of biologically interesting compounds can and have been synthesized with positron-emitting radionuclide tags (especially 11C and 18F) for PET imaging studies (Fowler and Wolf 1986; Tewson and Krohn 1998). These compounds include those that measure blood flow, substrate metabolism, protein synthesis, receptor binding, enzyme activity, and gene expression. Direct labeling of drugs, both therapeutic and drugs of abuse, is also possible. The flexibility of labeling any molecule that targets a specific biological pathway or marker and the ability to image the spatial distribution of the labeled molecule in vivo over time open up many potential uses for PET from basic research in animal models right up to the clinic.
The actual mass of tracer that must be injected to obtain an interpretable image depends on the specific activity with which the labeled compound can be synthesized. The specific activity, defined as the measured radioactivity of a sample divided by the mass of the sample, is determined by the fraction of molecules within a sample that are radiolabeled. The specific activity is typically on the order of 103 Ci/mmol for 11C and 18F, leading to concentrations of injected tracer that often are subnanomolar in humans and therefore below the levels generally required for pharmacological effects. In small animal studies, the mass of the tracer can in certain circumstances be a limiting factor in the study of receptor systems that are easily saturable and when a high level of receptor occupancy causes pharmacological effects (Hume et al. 1998). In the vast majority of cases, however, PET imaging is truly a tracer technique (injected mass in the range of 10-6-10-10 g) in which the material administered by injection does not perturb the biological system under study. Typical injected tracer doses required are in the 5- to 20-mCi range for human studies, 2- to 8-mCi range for nonhuman primate studies, 0.5- to 2-mCi range for rat studies, and 50- to 200-µCi range for studies in mice with currently available PET instrumentation. Radiation dose is a limiting factor in human studies but not a factor in animal studies, in which the mass and volume of the injected tracer and count-rate limitations of the PET scanner tend to determine the maximum injected dose. The radiation dose to the subject is highly dependent on the tracer and its pharmacokinetics.
Another important characteristic of PET imaging is that it enables the distribution of the radionuclide (and therefore the molecule to which it is attached) to be measured in quantitative units, assuming that appropriate corrections are made for physical factors such as gamma ray attenuation and scatter in the tissue. The quantitative nature of PET is largely independent of the thickness of the object and the depth of the source within the subject. This factor contrasts with optical imaging, in which light scatter makes accurate quantification and good spatial resolution at depth extremely challenging. Compared with nuclear imaging techniques that use single gamma ray-emitting radionuclides, PET also has at least one order of magnitude advantage in sensitivity (the number of detected events per unit of injected dose) because the direction of an incident gamma ray is defined electronically by the detection of the opposing gamma ray. In single gamma ray imaging, the direction must be defined by using a physical collimator that permits only gamma rays from predefined directions to strike the detectors and absorbs all other gamma rays. This dramatically reduces the detected gamma ray event rate per unit of injected dose of radioactivity. One disadvantage of PET relative to other techniques, however, is that all radionuclides lead to the same energy (two 511-KeV photon) emissions. Therefore, it is not possible to perform dual radionuclide studies simultaneously.
The data obtained from a dynamic PET study are a time series of volumetric images that reflect the distribution of the radionuclide within the imaging field of view over time. For small animals, this imaging field of view can often encompass the whole body. Thus, PET provides information on both the spatial and temporal distribution of the tracer in a single study. PET data can be further analyzed by applying tracer kinetic models (Huang and Phelps 1986). When combined with arterial blood samples that measure tracer delivery to the tissues, these models can be used to calculate biologically important information such as the rates of accumulation, metabolism, and clearance of the labeled molecules, or the binding characteristics of the molecule with the target of interest, or all of this information. This modeling can be important in various applications because sometimes the total radioactivity at a given location (often calculated from the image as percentage of injected dose per gram of tissue) is not a sufficient measure of the kinetics of the underlying process.
PET Scanners for Animal ResearchPET scanners for human imaging have improved dramatically since their introduction in the mid-1970s. At the time of this writing, they consist of large numbers of detector elements (typically 10,000-20,000) arranged in multiple rings to provide coverage of 10 to 25 cm in the axial direction (Figure 2a). Clinical PET systems typically produce reconstructed tomographic images with a spatial resolution in the 5- to 8-mm range for the brain and the 8- to 15-mm range for the rest of the body. These resolutions result in volume resolution elements (which can be interpreted as the volume of tissue from which independent information can be extracted) of, at best, 0.1 cc and, more typically, 0.5 to 2 cc. A typical study of the human brain from a clinical PET scanner is shown in Figure 2b. These clinical systems have also been widely used for animal research, predominantly in the larger laboratory animals such as nonhuman primates (brain), dogs (heart), and pigs (heart), where the spatial resolution is often adequate. There have also been attempts to image smaller animals such as rats on these clinical scanners (Ingvar et al. 1991), but the spatial resolution is sufficient for only a narrow range of applications.
The first dedicated animal PET scanners were also designed for larger research animals, particularly for brain imaging in nonhuman primates (Cutler et al. 1992; Watanabe et al. 1992). These systems were designed to obtain somewhat higher spatial resolution than clinical PET systems. In addition, the systems developed by Hamamatsu (Hamamatsu K.K., Japan) employ a sophisticated tilt mechanism in the gantry, which allows the detector ring to be placed horizontally, thus enabling PET imaging of conscious monkeys trained to sit in a chair (Tsukada et al. 2000). Among the several dedicated primate PET systems in existence, the latest generation high-resolution Hamamatsu system (SHR-7700) (Figure 2c) provides approximately 2.6-mm resolution in the transaxial directions and 3.2-mm resolution in the axial direction, yielding a volume resolution element of 22 mm3 or 0.022 cc (Watanabe et al. 1997). An example of a study from this system is shown in Figure 2d.
With the tremendous advances in mouse genomics and the wide range of animal models of human disease based on smaller laboratory animals such as mice and rats, there has been significant motivation to extend PET techniques to imaging of small animals. Several key challenges are involved. First, the spatial resolution must be improved substantially to enable individual organs and large substructures of organs to be identified reliably by PET. This improvement is necessary if PET is to be useful for a broad range of applications in small animals, beyond special cases such as imaging of large implanted tumors, and the imaging of specific receptor systems that are highly localized (e.g., the dopaminergic system). Second, the technology must evolve to a point where the PET scanner becomes a compact, relatively low-cost, high-throughput, and easily used benchtop system. Finally, PET radiopharmaceuticals must be widely available at low cost without the need for every laboratory to operate its own cyclotron. The first two challenges, which relate directly to the detector technology and scanner design, are being addressed by several research groups and companies worldwide at the time of this writing.
The first dedicated small animal PET scanner was developed through a collaboration between the Medical Research Council Cyclotron Unit, Hammersmith (London, England) and CTI PET Systems Inc. (Knoxville, Tennessee) (Bloomfield et al. 1995, 1997). This system took the standard detector technology being developed for clinical PET systems but placed these detector units in a small 12-cm-diameter ring to form a compact PET system for rodent imaging. Although the spatial resolution was not superior to that found in clinical systems, it did elegantly demonstrate the concept of a dedicated small animal PET system. Even with the spatial resolution limitations, a tremendous amount of useful research was carried out with this system in the rat brain, by focusing on targets (e.g., the striatum) that could be visualized with PET tracers specific to the dopaminergic system.
In the mid-1990s, numerous groups began to develop small animal PET scanners with detector technology developed specifically for that application and with improved spatial resolution. One of the first completed systems was developed at Sherbrooke University. This PET scanner was the first to use solid state photon detectors, called avalanche photodiodes, to read out the scintillator crystals in place of relatively bulky photomultiplier tubes (Lecomte et al. 1996; Marriott et al. 1994). Another system, microPET, was developed in our laboratory at UCLA and was based on a new scintillator material, lutetium oxyorthosilicate, coupled by optical fibers to multichannel photomultiplier tubes (Figure 2e) (Chatziioannou et al. 1999; Cherry et al. 1997). This system, which has a volumetric resolution of 0.006 cc, has been used to study more than 3000 research animals between 1996 and 2000 (Figure 2f). A commercial version of this system is now being manufactured by Concorde Microsystems Inc. (Knoxville). Several other PET systems dedicated to small animal imaging have been developed (Del Guerra et al. 1998; Pichler et al. 1998; Tavernier et al. 1995; Weber et al. 1997). Most of these systems are based on some form of scintillation detector, although multiwire proportional chamber detector technology has also been applied to small animal PET. A very high-level resolution system based on multiwire proportional chamber technology is manufactured by Oxford Positron Systems (Weston-on-the-Green, England) (Jeavons et al. 1999). The typical image resolution of many of these systems is in the 1- to 2.5-mm range and yields volumetric resolutions in the range of 0.001 to 0.015 cc.
Physical constraints on the ultimate spatial resolution attainable with PET exist and are caused by the range the positron travels before annihilation with an electron (positron range) in addition to the fact that the two gamma rays are not emitted exactly 180° apart (noncolinearity) due to the residual momentum of the electron and positron at annihilation. These two factors, combined with detector resolution, sampling, and the statistical quality of the data that can be achieved for a reasonable injected dose, will probably limit the resolution of PET images in rodents to approximately a 0.5- to 1-mm range (0.001-0.0001 cc volumetrically) with low energy positron emitters such as 11C and 18F. To reach submillimeter resolution reliably in PET images of small animals, substantial future improvements in the spatial resolution and the overall sensitivity of animal PET scanners are required.
Applications of PET in Animal Research
Extensive literature exists on the use of PET in animal research, ranging from the mouse up to the monkey. It is not possible to review all of the papers in this field comprehensively. The purpose of this section is therefore to provide an overview of representative work that demonstrates some of the different areas in which PET imaging has been applied in animal research.
One of the major research roles of PET has been in neuroreceptor imaging. This work now spans several species, from mouse to man. Some of the early studies in nonhuman primates were directed primarily at evaluating and validating new PET ligands for eventual human use (e.g., Wong et al. 1993; Yousef et al. 1996). The use of PET neuroreceptor imaging extends to testing of new drugs targeted to specific receptor systems (Burns et al. 1999; Fowler et al. 1999) to the study of drugs of abuse such as cocaine (Tsukada et al. 1996; Volkow et al. 1995) and methamphetamine (Melega et al. 1997). Some studies also use PET to monitor outcome longitudinally after striatal grafts in models of Parkinson's disease (Brownell et al. 1998; Fricker et al. 1997; Torres et al. 1995) and to evaluate the effects of neuroprotective and neurotrophic factors (Melega et al. 2000) in neurodegenerative disease and substance abuse. The majority of these studies focus on the dopaminergic system because there are several PET ligands for different components of this system and the striata are easily visualized by PET, even in the mouse (Chatziioannou et al. 1999).
A glucose analog ([18F]2-fluoro-2-deoxy-D-glucose [FDG1]) is the most frequently used clinical PET tracer and has also been used widely in animal studies. Uptake of this PET probe is reflective of regional glucose metabolism and is altered in a wide variety of disorders and disease states (Phelps 2000). FDG PET studies have been used to track metabolic development in the monkey brain (Moore et al. 2000a), to study events after traumatic brain injury in the rat (Moore et al. 2000b), and to study brain plasticity after hemidecortication (Kornblum et al. 2000). FDG has also been used to measure myocardial metabolism in the dog heart, and the methods are now being extended to the rat myocardium (Kudo et al. 1999b; Lapointe et al. 1999a). In conjunction with measures of blood flow (Kudo et al. 1999a), PET can be used to study the balance between flow and metabolism in chronic models of coronary artery disease. Some small animal PET scanners permit gating of the heart, which allows systolic and diastolic myocardial images of the rat heart to be obtained (Lapointe et al. 1999a). FDG is also widely used in cancer studies because cancer cells reveal increased accumulation of this tracer, probably due to increased glucose utilization and increased expression of the Glut-1 transporter (Brown et al. 1996). FDG has been used in mouse and rat tumor models, and one study has used PET longitudinally to monitor the effects of photodynamic therapy on implanted tumors (Lapointe et al. 1999b).
The applications mentioned above are critically dependent on radiolabeled tracers to study a specific process. One of the key limitations of the use of PET in animal research is that the synthesis of new probes can take many months to years to develop and optimize because the relatively short-lived isotopes available for PET generally make nuclear chemistry the rate-limiting step for most applications. In many cases, autoradiography and the use of beta-emitters (e.g., 3H, 14C) can help accelerate the testing of new probes for eventual use as PET probes (by changing the label to a positron emitter). It is necessary to validate that the final PET probe is stable in vivo and that the isotope does not dissociate from the parent molecule. The development of universal probes that could monitor many different molecular processes would greatly speed up animal research. In fact, several potential categories of universal probes are possible and are discussed below.
If one can synthesize a small segment of DNA and radiolabel it at one end (referred to as radioactive antisense oligodeoxynucleotide or RASON1), then one has the potential to build a universal PET tracer (Figure 3). This potential exists because the labeling process can be made independent of the exact sequence of the DNA. The sequence of the DNA determines what process the probe will measure. The sequence can be chosen to be complimentary to the messenger RNA (mRNA) sequence of a given gene. If the gene of interest is expressed, than the RASON can hybridize (bind) to the mRNA of that gene and specific signal can be detected with the PET camera. In those cells that do not express the gene, the probe can still enter the cell but not be significantly retained because the mRNA is not present. The holy grail of PET and many other imaging modalities would be such a universal probe. This probe could easily be synthesized in a DNA synthesizer and be rapidly radiolabeled independent of the exact sequence. Ultimately, this approach may be highly generalizable to the genes that produce sufficient levels of mRNA. Several factors currently limit the use of this approach, which is being investigated by numerous investigators (reviewed by Gambhir 1999). These factors include (1) stabilizing the RASON against degradation from nucleases in the blood and within the cell, (2) delivering enough RASON across the cell membrane, (3) allowing the RASON to hybridize to the mRNA target within the constraints of the isotope half-life, (4) synthesizing RASON probes with a high enough specific activity (typically 1000-10,000 Ci/mmol) to detect relatively low levels of mRNA, (5) minimizing the interaction of RASON probes with intracellular proteins and nontarget mRNAs so that unbound RASONs can efflux from the cell rapidly. Nevertheless, all of the issues listed above, and especially the last, have limited the use of the RASON as a universal probe. It should be noted that because this approach tracks mRNA levels, some applications in which it is critical to assay levels of protein or receptor as opposed to their corresponding mRNA will not be ideally suited for such a RASON based approach. It is likely that within the next decade that this approach or a modification of it will become reality.
A second universal approach is more useful in animals than in humans because its operating success depends on manipulation of the cellular genome. This approach is commonly referred to as a reporter gene approach. The basic concept (Figure 4) is to alter a particular set of cells to express a gene that would normally not be expressed and one that is not expressed elsewhere in the animal (Gambhir et al. 1999a,b). The gene product leads to an enzyme or receptor that is capable of trapping a radiolabeled probe. Several features make this approach quite useful to study many different processes. The promoter that drives the transcription of the reporter gene can be changed easily. One can use the same sequence as a promoter that is from any endogenous gene and couple it to the reporter gene to construct a chimeric gene. This allows one to monitor the endogenous gene indirectly because the same promoter sequence drives the transcription of both the endogenous and reporter genes. Therefore, it is not necessary to develop a probe for each new gene but instead, cells can simply carry the chimeric gene consisting of the endogenous gene promoter and the reporter gene sequence. Alternatively, one can drive the transcription of the reporter gene using a constitutive promoter (that is always active). In this case, when cells carry the chimeric gene consisting of the constitutive promoter driving the reporter gene, they will trap reporter probe. This process is useful for monitoring cells as they migrate from one part of the body to another.
The reporter gene approach is limited in that the reporter gene must be introduced into target tissue(s). This introduction can be accomplished in a variety of ways, including viruses or nonviral constructs that carry the reporter gene. Alternatively, it is possible to construct transgenic mice that carry the reporter gene in every cell but express it only in cells in which transcription is activated from a user-chosen promoter. A huge increase in the number of transgenic mice that carry imaging reporter genes is likely because of the potential to study many different processes noninvasively, using imaging technologies like microPET.
Specific examples of two reporter genes for PET imaging are the herpes simplex type 1 virus thymidine kinase (HSV1-tk1) (Gambhir et al. 1998, 1999b, 2000a; Tjuvajev et al. 1999a) and the dopamine type 2 receptor (D2R1) (MacLaren et al. 1999). The HSV1-tk reporter gene encodes for the enzyme denoted as HSV1-TK. This enzyme is capable of phosphorylating various substrates (e.g., ganciclovir, penciclovir), thereby trapping them in a given cell. Cells that do not express HSV1-tk do not significantly trap the reporter probes, and therefore a very low level of background signal is present. The natural function of this enzyme is to phosphorylate thymidine; however, because HSV1-TK has relaxed substrate specificity, it also phosphorylates other probes. At the time of this writing, one of the most effective imaging probes is side-chain fluorine-18 labeled penciclovir (FHBG1). Because HSV1-tk is a foreign gene, it can produce an immune response in immunocompetent mice. This is not an issue in transgenic mice because the mice have carried this gene while their immune system matures. FHBG does not cross the blood brain barrier so that this approach may not be useful if the blood brain barrier is intact. FHBG that is not sequestered in cells expressing HSV1-tk is cleared by the hepatobiliary and renal systems. This clearance leads to background signal in the gall-bladder, gastrointestinal tract, kidneys, and bladder. The pharmacokinetics are rapid, with greater than 80% of the tracer cleared from blood within 60 min in mice.
The D2R reporter gene uses the fluorine-18 labeled reporter probe 3-(2'-18F-fluoroethyl)-spiperone, which binds to the D2R receptors. Although D2R is expressed in the brain striatum, there is not significant expression elsewhere leading to specific signal only in those tissues in which the D2R reporter gene has been activated. 3-(2'-18F-fluoroethyl)-spiperone is also cleared by the hepatobiliary and renal system. Together the HSV1-tk and D2R reporter genes allow for monitoring two separate molecular processes by coupling their expression to two different promoters. An example of a transgenic mouse carrying the HSV1-tk reporter gene driven by the albumin promoter imaged on a microPET is shown in Figure 5 (Herschman et al. 2000).
Applications of the PET reporter genes are rapidly growing and include the following: (1) monitoring cell trafficking as cells metastasize to different organs; (2) monitoring endogenous gene expression indirectly through a reporter gene in transgenic animals carrying the PET reporter gene; (3) studying the interaction of two population of cells (e.g., immune system and tumor cells) over time; and (4) studying gene delivery and expression in living animals to optimize the delivery of a specific gene to a chosen target. It is noteworthy that for gene therapy approaches, methods have already been developed to link the expression of the PET reporter gene to any arbitrary therapeutic gene (Tjuvajev et al. 1999b; Yu et al. 2000). This development should lead to methods to monitor therapeutic gene expression in living animals and eventually humans, thereby greatly helping the field of gene therapy. Many additional applications exist but are beyond the scope of this review.
Reporter approaches that utilize optical reporter genes are also quite powerful (Contag et al. 1998). The luciferase and green-fluorescent protein reporter genes are two examples that can be used in the same ways as the PET reporter genes. For luciferase, a substrate (luciferin) is administered by injection into the peritoneum; and when cells expressing luciferase accumulate the luciferin, light is produced by using ATP as an energy source. For the green-fluorescent protein, an input wavelength must be provided and an output wavelength of light is produced (Yang et al. 2000). These optical approaches have the distinct advantage of low background signal, ease of use, and low cost; however, they are limited by signal scatter and absorption limiting studies deep in tissues, and they lack direct extension to most human applications. We are currently developing reporter systems that are capable of being imaged using CCD optical and small animal PET technologies by coupling the two reporter genes in a single vector.
Other general approaches for probe development include radiolabeled antibody fragments that can target receptors on the cell surface (Wu and Yazaki 2000). Antibodies can be rapidly generated against most target proteins and usually labeled in the same way regardless of the exact amino acids that compose the antibody. The selection of small fragments that bind to the target molecule can be used to optimize the clearance of the molecule from tissues that do not express the target. A recent example of this approach is the use of copper-64 radiolabeled antibody fragment targeted to the carcinoembryonic antigen receptor in a tumor xenograft model (Wu et al. 2000).
Future Challenges
PET appears to be developing rapidly as a viable tool for animal research. It allows noninvasive measurement of many different biological processes and permits longitudinal calculations of animal models. High-resolution dedicated animal PET scanners are available, and additional improvements in performance, cost, and ease of use can be anticipated. Nevertheless, several challenging issues still must be addressed for PET to become a routine part of the biological sciences.
The first challenge relates to the necessity of immobilizing the animal during study. Such immobilization is accomplished using general anesthetics, which are well known to have significant effects on the cardiovascular, respiratory, and central nervous systems (e.g., Gjedde and Rasmussen 1980; La Manna and Harik 1986). Careful consideration should be given to the choice of anesthetic to study a particular process so that effects are minimized. The consistency of depth of anesthesia during a study and from one study to another is also important. Careful control of anesthesia is also necessary to ensure long-term survival of the animals in longitudinal studies, in which the same animal may be anesthetized and scanned weekly for 6, 8, or even 10 wk. Although immobilization of an animal in a PET scanner is possible using restraining devices, it leads to extremely high stress levels, which can create conditions as poorly reflective of normal physiology as those of an anesthetized animal (Lasbennes et al. 1986). It has been possible to train nonhuman primates to undergo PET scans while conscious, although the effort to train these animals is significant (Tsukada et al. 2000).
To perform fully quantitative PET studies, information on the time course of tracer delivery to the tissue is required. This "input function" is best obtained from direct arterial blood sampling during the PET study. This function becomes extremely challenging in the smaller laboratory animals, particularly in longitudinal studies in which blood samples are required from the same animal repeatedly. It has been possible to perform repeated blood sampling in rat studies (Lapointe et al. 1998; Moore et al. 2000b), but in the mouse, the small size of the vessels and the tiny total blood volume make direct blood sampling extremely difficult. Alternatives include imaging the activity in the left ventricular blood pool to estimate the input function, or using tissue reference regions, where there is no specific retention of the probe (Green et al. 1998; Lammertsma and Hume 1996).
The final major challenge to the widespread use of animal PET relates to cost and tracer access. Sites with an existing clinical PET program and a radiochemistry development group can start an animal PET program for only the capital cost of the scanner. Others will face the choice of buying a cyclotron and starting their own radiochemistry program, or using the restricted set of 18F-labeled PET tracers that are used clinically and are available through commercial radionuclide distribution centers and making use of the longer lived or generator-produced PET radionuclides shown in Table 1. Fortunately, the number of sites with biomedical cyclotrons continues to grow, and the rapid growth in clinical PET is fueling the need for PET radiopharmaceuticals, which is being met with an ever-expanding network of regional distribution centers. Access to PET tracers is now easier than it has ever been, and the range of tracers that can be delivered to the door is likely to grow quickly in the next 5 to 10 yr.
Another likely important development is that of multimodality systems for small animal imaging. PET is powerful because of its high-level sensitivity, but it lacks anatomical information. Through the use of small animal systems such as microcomputerized axial tomography, it is possible to obtain detailed anatomical information at ~100 micron resolution (Paulus et al. 2000). Systems that combine microPET and microcomputerized axial tomography technology are under development, and an animal can be imaged in each scanner sequentially with software-based approaches to fuse the image sets. Additionally, by building vectors compatible with both optical CCD and PET technologies, it should be possible to image an animal in both an optical and a PET system. This approach would be powerful because a mouse could be screened rapidly and easily with the optical CCD system and then imaged in the microPET when detailed tomographic information is desired. Reporter genes that combine both optical and PET reporter genes are currently under construction and should help facilitate multimodality imaging.
Although the challenges remain significant, the field of animal PET is growing rapidly. By the end of 2001, it is estimated that 30 sites will have some form of dedicated animal PET scanner. If additional innovation can improve the performance and reduce the complexity and cost of animal PET systems, it is possible that this technology will spread widely into major research institutions and the pharmaceutical and biotechnology industries. The future appears bright, although much work remains to refine, simplify, validate fully, and evaluate PET for animal research studies.
Acknowledgments
The authors thank their many colleagues and collaborators at the University of California Los Angeles (UCLA) for useful discussions, and Dr. Takaji Yamashita, Hamamatsu Photonics K.K., for providing images and photographs for this manuscript. The work described in this paper was supported in part by National Institutes of Health (NIH) grants RO1 CA69370, NIH RO1 CA82214-01, NIH PO1 1P50CA86306-01, the UCLA-Jonsson Comprehensive Cancer Center, and DOE contract DE-FC03-87-ER60615.
1Abbreviations used in this article: CCD, charged-coupled device; D2R, dopamine type 2 receptor; FDG, [18F]2-fluoro-2-deoxy-D-glucose ; FHBG, fluorine-18 labeled penciclovir; HSV1-tk, herpes simplex virus type 1 thymidine kinase ; PET, positron emission tomography; RASON, radioactive antisense oligodeoxynucleotide.
References
Bloomfield PM, Myers R, Hume SP, Spinks TJ, Lammertsma AA, Jones T. 1997. Three-dimensional performance of a small-diameter positron emission tomograph. Phys Med Biol 42:389-400.
Bloomfield PM, Rajeswaran S, Spinks TJ, Hume SP, Myers R, Ashworth S, Clifford KM, Jones WF, Byars LG, Young J, Andreaco M, Williams CW, Lammertsma A, Jones T. 1995. The design and physical characteristics of a small animal positron emission tomograph. Phys Med Biol 40:1105-1126.
Brown RS, Leung JY, Fisher SJ, Frey KA, Ethier SP, Wahl RL. 1996. Intratumoral distribution of tritiated-FDG in breast carcinoma: Correlation between Glut-1 expression and FDG uptake. J Nucl Med 37:1042-1047.
Brownell AL, Livni E, Galpern W, Isacson O. 1998. In vivo PET imaging in rat of dopamine terminals reveals functional neural transplants. Ann Neurol 43:387-390.
Burns HD, Hamill TG, Eng WS, Francis B, Fioravanti C, Gibson RE. 1999. Positron emission tomography neuroreceptor imaging as a tool in drug discovery, research and development. Curr Opin Chem Biol 3:388-394.
Chatziioannou AF, Cherry SR, Shao Y, Silverman RW, Meadors K, Farquhar TH, Pedarsani M, Phelps ME. 1999. Performance evaluation of microPET: A high-resolution lutetium oxyorthosilicate PET scanner for animal imaging. J Nucl Med 40:1164-1175.
Cherry SR, Shao Y, Silverman RW, Meadors K, Siegel S, Chatziioannou A, Young JW, Jones W, Moyers JC, Newport D, Andreaco M, Paulus M, Binkley D, Nutt R, Phelps ME. 1997. MicroPET: A high resolution PET scanner for imaging small animals. IEEE Trans Nucl Sci 44:1161-1166.
Contag PR, Olomu IN, Stevenson DK, Contag CH. 1998. Bioluminescent indicators in living mammals. Nat Med 4:245-247.
Cutler PD, Cherry SR, Hoffman EJ, Digby WM, Phelps ME. 1992. Design features and performance of a PET system for animal research. J Nucl Med 33:595-604.
Del Guerra A, Di Domenico G, Scandola M, Zavattini G. 1998. YAP-PET: First results of a small animal positron emission tomograph based on YAP:Ce finger crystals. IEEE Trans Nucl Sci 45:3105-3108.
Fowler JS, Volkow ND, Wang GJ, Ding YS, Dewey SL. 1999. PET and drug research and development. J Nucl Med 40:1154-1163.
Fowler JS, Wolf AP. 1986. Positron emitter-labeled compounds: Priorities and problems. In: Phelps M, Mazziotta J, Schelbert H, eds. Positron Emission Tomography and Autoradiography. New York: Raven Press.
Fricker RA, Torres EM, Hume SP, Myers R, Opacka-Juffrey J, Ashworth S, Brooks DJ, Dunnett SB. 1997. The effects of donor stage on the survival and function of embryonic striatal grafts in the adult rat brain. II. Correlation between positron emission tomography and reaching behaviour. Neuroscience 79:711-721.
Gambhir SS. 1999. Imaging gene expression: Concepts and future outlook. In: Schiepers C, ed. Diagnostic Nuclear Medicine. Berlin: Springer. p 253-271.
Gambhir SS, Barrio JR, Herschman HR, Phelps ME. 1999a. Imaging gene expression: principles and assays. J Nucl Cardiol 6:219-233.
Gambhir SS, Barrio JR, Phelps ME, Iyer M, Namavari M, Satyamurthy N, Wu L, Green LA, Bauer E, MacLaren DC, Nguyen K, Berk AJ, Cherry SR, Herschman HR. 1999b. Imaging adenoviral-directed reporter gene expression in living animals with positron emission tomography. Proc Natl Acad Sci U S A 96:2333-2338.
Gambhir SS, Barrio JR, Wu L, Iyer M, Namavari M, Satyamurthy N, Bauer E, Parrish C, MacLaren DC, Borghei AR, Green LA, Sharfstein A, Berk AJ, Cherry SR, Phelps ME, Herschman HR. 1998. Imaging of adenoviral-directed herpes simplex virus type 1 thymidine kinase reporter gene expression in mice with radiolabeled ganciclovir. J Nucl Med 39:2003-2011.
Gambhir SS, Bauer E, Black ME, Liang Q, Kokoris MS, Barrio JR, Iyer M, Namavari M, Phelps ME, Herschman HR. 2000a. A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity for imaging reporter gene expression with positron emission tomography. Proc Natl Acad Sci U S A 97:2785-2790.
Gambhir SS, Herschman HR, Cherry SR, Barrio JR, Satyamurthy N, Toyokuni T, Phelps ME, Larson SM, Balatoni J, Finn R, Sadelain M, Tjuvajev J, Blasberg R. 2000b. Imaging transgene expression with radionuclide imaging technologies. Neoplasia 2:118-136.
Gjedde A, Rasmussen M. 1980. Pentobarbital anesthesia reduces blood-brain glucose transfer in the rat. J Neurochem 35:1382-1387.
Green LA, Gambhir SS, Srinivasan A, Banerjee PK, Hoh CK, Cherry SR, Sharfstein S, Barrio JR, Herschman HR, Phelps ME. 1998. Noninvasive methods for quantitating blood time-activity curves from mouse PET images obtained with fluorine-18-fluorodeoxyglucose. J Nucl Med 39:729-734.
Herschman HR, MacLaren DC, Iyer M, Namavari M, Bobinski K, Green LA, Wu L, Berk AJ, Toyokuni T, Barrio JR, Cherry SR, Phelps ME, Sandgren EP, Gambhir SS. 2000. Seeing is believing: Non-invasive, quantitative and repetitive imaging of reporter gene expression in living animals, using positron emission tomography. J Neurosci Res 59:699-705.
Huang S-C, Phelps ME. 1986. Principles of tracer kinetic modeling in positron emission tomography and autoradiography. In: Phelps M, Mazziotta J, Schelbert H, eds. Positron Emission Tomography and Autoradiography. New York: Raven Press. p 287-346.
Hume SP, Gunn RN, Jones T. 1998. Pharmacological constraints associated with positron emission tomographic scanning of small laboratory animals. Eur J Nucl Med 25:173-176.
Ingvar M, Eriksson L, Rogers GA, Stone-Elander S, Widén L. 1991. Rapid feasibility studies of tracers for positron emission tomography: High-resolution PET in small animals with kinetic analysis. J Cereb Blood Flow Metab 11:926-931.
Jeavons AP, Chandler RA, Dettmar CAR. 1999. A 3D HIDAC-PET camera with sub-millimetre resolution for imaging small animals. IEEE Trans Nucl Sci 46:468-473.
Jones T. 1996. The imaging science of positron emission tomography. Eur J Nucl Med 23:807-813.
Knapp FF, Mirzadeh S. 1994. The continuing important role of radionuclide generator systems for nuclear medicine. Eur J Nucl Med 21:1151-1165.
Kornblum HI, Araujo DM, Annala AJ, Tatsukawa KJ, Phelps ME, Cherry SR. 2000. In vivo imaging of neuronal activation and plasticity in the rat brain with microPET, a novel high-resolution positron emission tomograph. Nat Biotechnol 18:655-660.
Kudo T, Annala AJ, Cherry SR, Phelps ME, Schelbert HR. 1999a. Measurement of myocardial blood flow during occlusion/reperfusion in rats with dynamic microPET imaging. (Abstract). J Nucl Med 40:6P.
Kudo T, Annala AJ, Cherry SR, Phelps ME, Schelbert HR. 1999b. Noninvasive measurement of F-18 deoxyglucose concentrations in rat myocardium with UCLA microPET. (Abstract). J Nucl Med 40:183P.
La Manna JC, Harik SI. 1986. Regional studies of blood-brain barrier transport of glucose and leucine in awake and anesthetized rats. J Cereb Blood Flow Metab 6:717-723.
Lammertsma AA, Hume SP. 1996. Simplified reference tissue model for PET receptor studies. Neuroimage 4(Pt 1):153-158.
Lapointe D, Bentourkia M, Cadorette J, Rodrique S, Ouellet R, Benard F, Van Lier JE, Lecomte R. 1999a. High-resolution cardiac PET in rats. (Abstract). J Nucl Med 40:185P.
Lapointe D, Brasseur N, Cadorette J, La Madeleine C, Rodrigue S, van Lier JE, Lecomte R. 1999b. High-resolution PET imaging for in vivo monitoring of tumor response after photodynamic therapy in mice. J Nucl Med 40:876-882.
Lapointe D, Cadorette J, Rodrigue S, Rouleau D, Lecomte R. 1998. A microvolumetric blood counter/sampler for metabolic PET studies in small animals. IEEE Trans Nucl Sci 45:2195-2199.
Lasbennes F, Lestage P, Bobillier P, Seylaz J. 1986. Stress and local cerebral blood flow: studies on restrained and unrestrained rats. Exp Brain Res 63:163-168.
Lecomte R, Cadorette J, Rodrigue S, Lapointe D, Rouleau D, Bentourkia M, Yao R, Msaki P. 1996. Initial results from the Sherbrooke avalanche photodiode positron tomograph. IEEE Transactions on Nuclear Science 43:1952-1957.
MacLaren DC, Gambhir SS, Satyamurthy N, Barrio JR, Sharfstein S, Toyokuni T, Wu L, Berk AJ, Cherry SR, Phelps ME, Herschman HR. 1999. Repetitive, non-invasive imaging of the dopamine D2 receptor as a reporter gene in living animals. Gene Ther 6:785-791.
Marriott CJ, Cadorette JE, Lecomte R, Scasnar V, Rousseau J, van Lier JE. 1994. High-resolution PET imaging and quantitation of pharmaceutical biodistributions in a small animal using avalanche photodiode detectors. J Nucl Med 35:1390-1396.
McCarthy TJ, Welch MJ. 1998. The state of positron emitting radionuclide production in 1997. Semin Nucl Med 28:235-246.
Melega WP, Lacan G, Desalles AA, Phelps ME. 2000. Long-term methamphetamine-induced decreases of [(11)C]WIN 35,428 binding in striatum are reduced by GDNF: PET studies in the vervet monkey. Synapse 35:243-249.
Melega WP, Raleigh MJ, Stout DB, Lacan G, Huang SC, Phelps ME. 1997. Recovery of striatal dopamine function after acute amphetamine- and methamphetamine-induced neurotoxicity in the vervet monkey. Brain Res 766:113-120.
Moore AH, Hovda DA, Cherry SR, Villablanca JP, Pollack DB, Phelps ME. 2000a. Dynamic changes in cerebral glucose metabolism in conscious infant monkeys during the first year of life as measured by positron emission tomography. Dev Brain Res 120:141-150.
Moore AH, Osteen CL, Chatziioannou AF, Hovda DA, Cherry SR. 2000b. Quantitative assessment of longitudinal metabolic changes in vivo following traumatic brain injury in the adult rat using FDG-microPET. J Cereb Blood Flow Metab 20:1492-1501.
Ollinger JM, Fessler JA. 1997. Positron-Emission Tomography. IEEE Signal Processing Magazine. January:43-55.
Paulus MJ, Gleason SS, Kennel SJ, Hunsicker PR, Johnson DK. 2000. High resolution X-ray computed tomography: An emerging tool for small animal cancer research. Neoplasia 2:62-70.
Phelps ME. 2000. Positron emission tomography provides molecular imaging of biological processes. Proc Natl Acad Sci U S A 97:9226-9233.
Pichler B, Boning C, Lorenz E, Mirzoyan R, Pimpl W, Schwaiger M, Ziegler SI. 1998. Studies with a prototype high resolution PET scanner based on LSO-APD modules. IEEE Trans Nucl Sci 45:1298-1302.
Tavernier S, Bruyndonckx P, Debruyne J, Etienne L, Folger H, Hartmann W, Rajeswaran S, Smolik W, Van Lancker L, Zhang S. 1995. First results from a prototype PET scanner using BaF2 scintillator and photosensitive wire chambers. In: Trendler R, ed. Conference Record of the 1994 IEEE Nuclear Science Symposium and Medical Imaging Conference, vol 4. Piscataway NJ: IEEE Press. p 1885-1887.
Tewson TJ, Krohn KA. 1998. PET radiopharmaceuticals: State-of-the-art and future prospects. Semin Nucl Med 28:221-234.
Tjuvajev JG, Chen SH, Joshi A, Joshi R, Guo ZS, Balatoni J, Ballon D, Koutcher J, Finn R, Woo SL, Blasberg RG. 1999a. Imaging adenoviral-mediated herpes virus thymidine kinase gene transfer and expression in vivo. Cancer Res 59:5186-5193.
Tjuvajev JG, Joshi A, Callegari J, Lindsley L, Joshi R, Balatoni J, Finn R, Larson SM, Sadelain M, Blasberg RG. 1999b. A general approach to the non-invasive imaging of transgenes using cis-linked herpes simplex virus thymidine kinase. Neoplasia 1:315-320.
Torres EM, Fricker RA, Hume SP, Myers R, Opacka-Juffry J, Ashworth S, Brooks DJ, Dunnett SB. 1995. Assessment of striatal graft viability in the rat in vivo using a small diameter PET scanner. Neuroreport 6:2017-2021.
Tsukada H, Harada N, Nishiyama S, Ohba H, Kakiuchi T. 2000. Dose-response and duration effects of acute administrations of cocaine and GBR12909 on dopamine synthesis and transporter in the conscious monkey brain: PET studies combined with microdialysis. Brain Res 860:141-148.
Tsukada H, Kreuter J, Maggos CE, Unterwald EM, Kakiuchi T, Nishiyama S, Futatsubashi M, Kreek MJ. 1996. Effects of binge pattern cocaine administration on dopamine D1 and D2 receptors in the rat brain: An in vivo study using positron emission tomography. J Neurosci 16:7670-7677.
Volkow ND, Fowler JS, Logan J, Gatley SJ, Dewey SL, MacGregor RR, Schlyer DJ, Pappas N, King P, Wang GJ. 1995. Carbon-11-cocaine binding compared at subpharmacological and pharmacological doses: A PET study. J Nucl Med 36:1289-1297.
Watanabe M, Okada H, Shimizu K, Omura T, Yoshikawa E, Kosugi T, Mori S, Yamashita T. 1997. A high resolution animal PET scanner using compact PS-PMT detectors. IEEE Trans Nucl Sci 47:1277-1282.
Watanabe M, Uchida H, Okada H, Shimizu K, Satoh N, Yoshikawa E, Ohmura T, Yamashita T, Tanaka E. 1992. A high resolution PET for animal studies. IEEE Trans Med Imag 11:577-580.
Weber D, Ivanovic M. 1999. Ultra-high-resolution imaging of small animals: Implications for preclinical and research studies. J Nucl Cardiol 6:332-344.
Weber S, Terstegge A, Engels R, Herzog H, Reinartz R, Reinhart P, Rongen F, Muller-Gartner HW, Halling H. 1997. The design of an animal PET: Flexible geometry for achieving optimal spatial resolution or high sensitivity. IEEE Trans Med Imag 16:684-689.
Weissleder R. 1999. Molecular imaging: Exploring the next frontier. Radiology 212:609-614.
Weissleder R, Tung C-H, Mahmood U, Bodganov A. 1999. In vivo imaging of tumors with protease-activated near-infrared fluorescent probes. Nat Biotech 17:375-378.
Wong DF, Yung B, Dannals RF, Shaya EK, Ravert HT, Chen CA, Chan B, Folio T, Scheffel U, Ricaurte GA, Neumeyer JL, Wagner HN, Kuhar MJ. 1993. In vivo imaging of baboon and human dopamine transporters by positron emission tomography using [11C]WIN 35,428. Synapse 15:130-142.
Wu AM, Yazaki PJ. 2000. Designer genes: Recombinant antibody fragments for biological imaging. Q J Nucl Med 44:268-283.
Wu AM, Yazaki PJ, Tsai S, Nguyen K, Anderson AL, McCarthy DW, Welch MJ, Shively JE, Williams LE, Raubitschek AA, Wong J, Toyokuni T, Phelps ME, Gambhir SS. 2000. High-resolution microPET imaging of carcinoembryonic antigen-positive xenografts by using a copper-64-labeled engineered antibody fragment. Proc Natl Acad Sci U S A 97:8495-8500.
Yang M, Baranov E, Jiang P, Sun FX, Li XM, Li L, Hasegawa S, Bouvet M, Al-Tuwaijri M, Chishima T, Shimada H, Moossa AR, Penman S, Hoffman RM. 2000. Whole-body optical imaging of green fluorescent protein-expressing tumors and metastases. Proc Natl Acad Sci U S A 97:1206-1211.
Yousef KA, Fowler JS, Volkow ND, Dewey SL, Shea C, Schlyer DJ, Gatley SJ, Logan J, Wolf AP. 1996. [18F]Haloperidol binding in baboon brain in vivo. Nucl Med Biol 23:47-52.
Yu Y, Annala AJ, Barrio JR, Toyokuni T, Satyamurthy N, Namavari M, Cherry SR, Phelps ME, Herschman HR, Gambhir SS. 2000. Quantitation of target gene expression by imaging reporter gene expression in living animals. Nat Med 6:933-937.
Figure 1 Basic physics of positron emission tomography (PET). A compound labeled with a positron-emitting radionuclide is introduced into the body usually by intravenous injection. When one of the radionuclide atoms decays, a positron is emitted, travels a very short distance in tissue (typically 0-1-100 mm for radionuclides of interest), and annihilates with an electron in the tissue. The mass of the two particles is converted into energy, which is emitted in the form of two back-to-back 511 keV gamma rays. A positron emission tomography scanner consists of a ring, or multiple rings, of gamma ray detectors that register simultaneous gamma ray hits and their location, thus defining the line along which the positron-emission took place. By collecting large numbers of gamma-ray pair events (typically 106 to 107) and using computed tomography methods, cross-sectional images reflecting the concentration of the positron-emitting radionuclide can be generated.
Figure 2 Montage reflecting the use of positron emission tomography (PET) from humans to rodent. In the top row, images represent PET scanners designed for human, monkey, and small animal imaging. The size of the scanners is approximately to scale. In the bottom row, brain images reflect glucose metabolism in humans (resolution ~5 mm), monkey (resolution ~2.5 mm), and rat (resolution ~1.5 mm) using the tracer [18F]2-fluoro-2-deoxy-D-glucose (FDG). The rat images are magnified by a factor of three for better visualization. a) HR+ clinical PET scanner (CTI/Siemens, Knoxville, Tennessee). b) PET images of the human brain obtained on the HR+ scanner after injection of FDG into a normal subject. Four transverse slices at different levels of the brain are shown. c) SHR-7700 animal PET scanner (Hamamatsu K.K., Japan). d) FDG PET images of a rhesus monkey obtained from the SHR-7700 scanner. Four transverse sections are shown. e) UCLA microPET scanner. f) FDG PET images of the rat brain obtained from the microPET scanner. Four coronal sections are shown.
Figure 3 Imaging with a radiolabeled antisense oligodeoxynucleotide (RASON). A small RASON molecule can be synthesized easily to target a particular portion of a mRNA specifically. If this probe can successfully enter cells, it can be used to detect levels of target mRNA through imaging. Nonspecific interaction of the RASON with other mRNA and proteins is also possible. In those cells in which the target mRNA is not expressed, the RASON would efflux back out of the cell. This approach is in early stages of development.
Figure 4 Principle of imaging with a reporter gene. A reporter gene driven by a promoter of choice can be delivered into target cells. If the reporter gene is expressed, the protein made (shown as spheres) can specifically trap a reporter probe. In those cells in which the reporter protein is not present, the reporter probe can enter the cell, efflux back out, and be cleared from the blood. The reporter gene can encode for a intracellular protein or a receptor that can be intracellular, on the cell surface, or both. This approach is a general approach for imaging gene expression because many different processes can be studied by using different promoters.
Figure 5 Micro-positron emission tomography (microPET) imaging of the herpes simplex virus type 1 thymidine kinase (HSV1-tk) reporter gene in a transgenic mouse. A transgenic mouse was developed in which in every cell of the mouse, the HSV1-tk reporter gene is present. However, because this reporter gene is driven by the albumin promoter, only cells in which albumin is expressed (predominantly hepatocytes in the liver) will lead to expression of HSV1-tk. The microPET tomographic image (coronal view) shown was obtained ~60 min after the injection of fluorine-18 labeled penciclovir (FHBG) (a probe specific for the HSV1-tk reporter gene product). There is specific signal predominantly from the liver, due to trapping of FHBG, and from the intestines, due to clearance of FHBG from the hepatobiliary system. FHBG was able to enter all cells (except for those within the brain) but was trapped only in the cells that expressed HSV1-tk. The scale on the right (in percentages of injected dose per gram [%ID/g]) is a measure of the amount of tracer in each region of the mouse.
Table 1 Positron-emitting radionuclides of interest
for biomedical studies
| Radionuclide | Half-life | Production |
11C | 20.3 min | Cyclotron |
Copyright © 2008. National Academy of Sciences.
All rights reserved.
500 Fifth St. N.W., Washington, D.C. 20001.
Terms of Use and Privacy Statement