Whole animal imaging

Abstract

Translational research plays a vital role in understanding the underlying pathophysiology of human diseases, and hence development of new diagnostic and therapeutic options for their management. After creating an animal disease model, pathophysiologic changes and effects of a therapeutic intervention on them are often evaluated on the animals using immunohistologic or imaging techniques. In contrast to the immunohistologic techniques, the imaging techniques are noninvasive and hence can be used to investigate the whole animal, oftentimes in a single exam which provides opportunities to perform longitudinal studies and dynamic imaging of the same subject, and hence minimizes the experimental variability, requirement for the number of animals, and the time to perform a given experiment. Whole animal imaging can be performed by a number of techniques including x-ray computed tomography, magnetic resonance imaging, ultrasound imaging, positron emission tomography, single photon emission computed tomography, fluorescence imaging, and bioluminescence imaging, among others. Individual imaging techniques provide different kinds of information regarding the structure, metabolism, and physiology of the animal. Each technique has its own strengths and weaknesses, and none serves every purpose of image acquisition from all regions of an animal. In this review, a broad overview of basic principles, available contrast mechanisms, applications, challenges, and future prospects of many imaging techniques employed for whole animal imaging is provided. Our main goal is to briefly describe the current state of art to researchers and advanced students with a strong background in the field of animal research.

In translational research, the gap between in vitro experiments and their clinical applications is often bridged by investigating the animal models of various human diseases. The pathophysiology of an animal and its response to a therapeutic intervention are often studied using one imaging technique or other. Most imaging techniques can be employed to perform transverse as well as longitudinal analyses of whole animals and this minimize the experimental variability, shorten the time required for a given experiment, and enable meaningful expenditure of research resources. Whole animal imaging can be performed using a number of imaging techniques. Depending upon their primary applications, the techniques for whole animal imaging can be broadly classified into two categories: structural and molecular imaging techniques. Structural imaging techniques primarily provide high-resolution images in millimeter to micrometer range and include ultrasound (US) imaging, X-ray computed tomography (CT), and magnetic resonance imaging (MRI). Image formation in these techniques is based upon the principle of spatial difference in interactions of the tissues with sound waves, X-rays, and radiofrequency (RF) electromagnetic fields, respectively. The techniques of the second category are primarily employed for molecular imaging and can again be classified into two subcategories on the basis of the principle of image formation; in many ways, however, MRI and US share properties of both structural and molecular imaging. The first subcategory comprises of nuclear imaging techniques such as positron emission tomography (PET) and single photon emission computed tomography (SPECT), where the fundamental basis for image formation is detection of radiation emitted by injected radioactive tracers. The second subcategory consists of optical imaging techniques such as bioluminescence imaging (BLI), fluorescence imaging (FLI), and optical coherent tomography (OCT). The fundamental basis for image formation in BLI and FLI is capturing of a light photon emitted by a living microorganism or a fluorophore, respectively, from inside the animal. OCT exploits the difference in optical scattering properties of individual tissues for image formation. Individual imaging techniques provide different kinds of information regarding structure, metabolism, and physiology of an animal. Information provided by the individual techniques can be synergized by combining the images provided by them using an image registration mechanism.^1,2

In this advanced review, an overview of the basic principle of image formation, contrast enhancement mechanisms, important applications, challenges, and future prospects of the techniques commonly employed for whole animal imaging is provided. Our main goal is to provide a brief description of the current state of art to the molecular biologists, biochemists, pathologists, geneticists, pharmacologists, and nuclear scientists who are working in the field of animal research. Information provided in this article would help these professionals in designing future experiments by considering the availability of these varied yet synergistic imaging techniques.

COMPUTED TOMOGRAPHY

Dedicated CT scanners employed for small animal imaging are a miniaturized version of clinical (human) CT scanners and are often called microcomputed tomogram (μCT) scanners.³ For μCT image acquisition, a number of X-ray beams of a known intensity are projected onto an imaging volume of interest from different rotation angles. The intensity of the X-rays passing through the volume is captured by an array of X-ray detectors placed around it. Multiple X-ray projections acquired at different rotation angles are combined to produce the final μCT image. The image contrast between different tissues in the volume is determined by the difference in their individual X-rays attenuation capacity.⁴ Knowledge of the intensity of the incident X-ray beam and that reaching the detectors allows the x-ray attenuating capacity of the tissue to be determined. Cardiorespiratory motion of the live animals during μCT image acquisition can be compensated by using prospective or retrospective respiratory and/or cardiac synchronization (i.e., gating).^5,6 A similar technique is digital subtraction angiography (DSA), in which an extracellular contrast agent is injected intravenously and data are acquired before and after contrast administration to generate angiographic images. A projection image is obtained to display the distribution of the contrast agent by subtracting the precontrast image (i.e., a mask) from the post-contrast image.⁴ Tomographic DSA has also been described in the literature to obtain 3D spatial information from DSA projections using tomosynthesis.⁷

Contrast Agents

The μCT contrast agents increase the degree of X-ray attenuation in the region of their distribution, and hence improve the μCT image contrast. Bolus injection of a clinically used iodine-based contrast agent such as Isovue-370 has been employed for small animal μCT imaging. Oftentimes, the volume of agent injected, the heart rate, blood flow, short circulation time, and the physiologic efficiency of renal excretion limit the amount of time the agents reside in small animal tissue. Therefore, it is sometimes difficult to obtain a sufficient number of projections for μCT imaging following a bolus injection of these agents.⁸ Constant infusion of these contrast agents have been employed as an alternative to the bolus injection.⁶

New contrast agents with a relatively long vascular half-life are now available. Two iodinated triglycerides (glyceryl-2-oleoyl-1, 3-di-7-(3-amino- 2,4,6-triiodophenyl)-heptanoate; also called ITG-DHOG), Fenestra-VC and Fenestra-LC (iodine content 50 mg/ml) are currently employed as hepatobiliary-selective blood-pool contrast agents.^9,10 Fenestra-VC is particularly useful for cardiac, tumor, and hepatobiliary imaging.^5,11,12 Because of its rapid uptake by hepatocytes, Fenestra-LC has a very short blood-pool contrast enhancement time; however, it has found applications for spleen and hepatobiliary imaging.¹³ Iodine-containing contrast agents are limited by the fact that they cannot be conjugated with biologic markers, which limits their utility for localization of specific targets.⁸ This limitation of the iodine-containing contrast agents can be overcome by using nanoparticle-based contrast agents which can also be conjugated with biomarkers. Various nanoparticle-based contrast agents with a relatively long half-life such as polymer-coated gold nanoparticles,¹⁴ gadolinium-coated gold nanoparticles,¹⁵ Polyethylene glycol (PEG)-coated colloidal gold nanoparticles,¹⁶ polymer-coated Bismuth-sulphide nanoparticles,¹⁷ and antibody specific nanoprobes⁸ have been developed for vascular imaging. Liposomal contrast agents have a long half-life¹⁸ and a high affinity to microvascular fenestrations.¹⁹ E-selectin-labeled liposomal contrast agents can be used to image activated endothelial cells.²⁰ Compounds such as tantalum²¹ and xenon²² have been employed as inhalation contrast agents.

Applications

μCT scanners can provide images at an almost cellular level of resolution³ and have been employed for cardiac, pulmonary, hepatic, splenic, renal, and tumor imaging and to study gene expression. Cardiac μCT images can be obtained with a temporal and spatial resolution of up to 100 μm and 10 ms, respectively, using prospective gating⁵ and up to 150 μm and 12 ms, respectively, using retrospective gating.⁶ Recently, a registration-based approach for 4D (three spatial dimensions and one temporal) cardiac μCT imaging using a combined prospective and retrospective gating approach has also been introduced.²³ Various cardiac parameters can be calculated from dynamic μCT images by analyzing the left ventricular volumes at different time points of the cardiac cycle.⁵ Dynamic μCT cardiac imaging may be useful for understanding various pathologic processes such as left ventricular wall thickening ⁵ and temporal evaluation of myocardial infarction-induced cardiac remodeling.²⁴ μCT imaging has also been employed to investigate cardiac phenotype and coronary artery patterning defects.^25,26

Small animal pulmonary μCT images with a spatial resolution as high as 91 μm and 150 μm have been obtained using prospective ²⁷ and retrospective ²² gating, respectively. μCT images acquired at different time points in the respiratory cycle can be used to calculate the respiratory lung volumes ²² and morphometric indices.^17,22,28,29 μCT imaging has been employed to characterize pulmonary airways, pulmonary vasculature,³⁰ microvascular permeability, and alveolar membrane edema,³¹ and for lung tumor analysis.^32,33 Small animal models of diseases such as emphysema ³⁴ and fibrosis ²⁴ have been investigated using μCT imaging.

μCT imaging of the liver and spleen is usually performed following contrast agent administration.^12,13,35 In vivo contrast-enhanced (CE) μCT images of liver with a spatial resolution as high as 44 μm have been produced and can be used to delineate tumors as small as 300 μm.^36,37 μCT imaging can also be used to delineate the hepatic vasculature ³⁸ and investigate the changes following ischemic/reperfusion injury.¹² Iodinated liposomes are particularly useful for splenic μCT imaging because of the fact that they get degraded in the spleen. Red pulp of the spleen appears more radio-intense than white pulp in CE-μCT images as a result of concentration of phagocytes in the former.³⁵

Renal μCT imaging is usually performed by in situ perfusion fixing and is particularly useful for glomerular analysis.^39–41 Conventional-source μCT scanners provide renal images with a spatial resolution in the range of 10–20 μm that can be used to calculate the total renal, cortical, and medullary volumes, glomerular diameters and numbers, and vascular volume fractions.^42,43 μCT images can also be used to visualize the 3D architecture of nephrons and renal blood vessels.^42,43

In vivo μCT imaging of murine proximal tibial metaphysis has been performed with a spatial resolution as high as 20 × 20 × 26 μm³.⁴⁴ However, most of the previous work for osteology applications of μCT imaging has been done via ex vivo analysis. In addition to bone mineral content and bone mineral density, μCT images can be employed to calculate morphometric indices.^44–47 Longitudinal bone survey can also be performed in order to investigate the bone growth and unloading-related changes.^44,48

μCT imaging is an important tool for longitudinal imaging of tumors (see Figure 1). μCT imaging has been used for hepatic ⁴⁹ and pulmonary ⁵⁰ tumor imaging. Gold nanoparticles conjugated with UM-A9 antibodies ⁸ and iodinated contrast agents such as Fenestra-VC ¹¹ have also been employed for CE-μCT tumor imaging. E-selectin specific μCT imaging can be used to image activated vessel walls ²⁰ and can be used for noninvasive monitoring of tumor angiogenesis.⁵¹ Multiple-contrast⁵² and negative-contrast⁵³ μCT colonography has also been devised to trace morphologic lesions and tumors in the murine colon in vivo. μCT imaging has also been employed to image the expression of sodium-iodide symporter (a reporter gene) in mouse tumors.⁵⁴ Perfusion DSA has been employed to investigate vasculature flow patterns.^55,56 Pulmonary blood volume, blood flow, and mean transit times can be calculated from DSA images^4,7,56 (see Figure 2). DSA has also been employed to study tumor angiogenesis.⁵⁷ Tomographic DSA enables functional perfusion measurements in 4D.⁷

FIGURE 1.

Virtual in vivo μCT images of growing lung tumors. Axial microcomputed tomogram (μCT) images of the thorax of a mouse at the indicated time points are shown. μCT images of two lung adenocarcinomas (a–c) and (d–f) and a lung adenoma (g–i) acquired at different ages. Tumors observed in the lung are circled. (Image reproduced with permission from Ref 33. Copyright 2008 Wiley-Blackwell).

FIGURE 2.

Sixteen successive images (starting at second heartbeat post-injection) from a digital subtraction angiography (DSA) sequence with a 100 ms contrast injection (A). The reproducibility for (n = 5) 100 ms injections (volume injected 83.22 μL) is shown by mean and standard error at each heartbeat (B) in a region of the aortic arch. Note that mean peak value of contrast-to-noise ratio (CNR) is 10. (Image reproduced with permission from Ref 7. Copyright 2007).

Challenges and Future Prospects

In order to understand human disease processes on the small animal models, acquired images need to be scaled according to dimensions and physiology of the animal;⁴ hence, smaller volume elements (i.e., voxels) are required. Yet, the signal-to-noise ratio in the images is dependent on the number of X-ray photons passing through the voxel. As such, the X-ray exposure dose in μCT imaging of small animals can be high when compared to that in human imaging. In fact, as one might predict, higher X-ray exposure dose is detrimental to life expectancy of live animals, may alter phenotypic expression of genetically engineered animals, and can cause radiation-induced neoplasia.³ Various approaches such as application of contrast agents, prospective gating, reduced X-ray scatter, monochromatic X-ray beams, fewer projections, alternate filtering, and beam modulation according to imaging cross section can be employed to decrease X-ray exposure dose.^5,58,59 Lastly, there are some daily practice challenges such as instrument availability, longer image acquisition times, and lower contrast agent availability that limit μCT utility.

The ability of μCT imaging to visualize the 3D architecture of basic functional units such as hepatic lobules and nephrons in vivo without physical disruption could help in further understanding the pathophysiology of various disease processes. Development of nanoparticles-based target-specific contrast agents would help in diagnosis of neoplastic lesions at the earlier stages. Newly developing refraction imaging techniques such as phase contrast imaging ^60,61 can enable high-quality image acquisition with a potential reduction in the radiation exposure by as much as 10,000 fold.⁵⁸

MAGNETIC RESONANCE IMAGING

Whole animal MRI can be performed on dedicated high-field MRI ⁶² and nuclear magnetic resonance (NMR) scanners as well as on clinical MRI scanners ⁶³ using dedicated receiving coils.⁶⁴ During MRI acquisition, nuclei with a nonzero magnetic moment produce a net magnetization when placed within a static magnetic field (B₀) and are then excited by applying an RF excitation pulse. Following the excitation, the nuclei reemit RF waves that are spatially localized by applying magnetic field gradients of known magnitude and orientation.⁶⁵ A variety of morphologic images with different contrasts can be acquired by selecting a suitable combination of imaging sequence parameters.^66,67 In addition, various techniques such as contrast-enhanced (CE) MRI, perfusion MRI, functional MRI (fMRI), diffusion tensor imaging (DTI), and MR spectroscopy (MRS) can also be used to extract different kinds of information. CE-MRI involves image acquisition following injection of a gadolinium-based contrast agent. Two important CE-MRI techniques, dynamic contrast-enhanced (DCE) MRI ⁶⁸ and dynamic susceptibility-contrast (DSC)-MRI⁶⁹ can be used to quantify contrast agent concentration by combining T1 (e.g., gadolinium chelates) and T2* (e.g., iron oxide particles) agents with MR imaging acquisitions sensitive to the effects of these specific agents. Perfusion MRI can be performed by any of the two methods, arterial spin labeling perfusion MRI (ASL-MRI) and DSC-MRI, and is used for perfusion quantification (regional blood flow and volume, etc.)⁷⁰ In ASL-MRI, the magnetization of a proton pool is labeled by applying a slice-selective RF pulse, and reduction of the signal in a T1–weighted image upon arrival of the magnetically labeled blood into the imaging volume is used for perfusion quantification.^71,72 A comparative analysis of ASL- and DSC-MRI techniques for perfusion MRI can be found elsewhere.⁷¹ In fMRI, the activity-related neuronal activity changes in specific brain regions are indirectly and noninvasively analyzed. The concentration of oxyhemoglobin increases in the vicinity of neuronal excitation resulting in a signal increase in blood-oxygen-level-dependent MRI images.⁷³ In DTI, the magnitude and direction of water diffusivity is analyzed by comparing multiple images acquired with gradient pulses applied in different directions.⁷⁴ In this way, it is possible to map fiber direction information (e.g., brain white matter tracks). The fundamental basis for MRS is the shift in resonance frequencies of various paramagnetic atoms such as ¹H, ³¹P, ¹³C, ¹⁹F, ¹⁷O, and ²³Na as a result of the electronic environment of a given nuclei; for example, electronegative nuclei such as oxygen deshield detectable hydrogen nuclei of their respective electrons providing a different chemical shift from other electron-rich hydrogen nuclei (e.g., fat). With these and many othermolecular level differences, MRS provides the platform to explore the biochemistry of these atoms in vivo.^75–77

Contrast Agents

Conventional MRI contrast agents influence the image contrast by reducing both the longitudinal (T1) as well as the transverse (T2) relaxation times of the water protons in their vicinity.⁷⁸ In this way, the contrast agents provide a positive and a negative contrast to the T1- and T2-weighted images, respectively. Detailed information regarding the mechanism of action, magnetic and chemical properties, and biodistribution pattern of various MRI contrast agents can be found elsewhere.⁷⁸ According to their magnetic properties, the MRI contrast agents can be classified as paramagnetic (e.g., gadolinium-based agents) or super-paramagnetic (e.g., iron oxide nanoparticles) agents.^78,79 Depending upon their biodistribution pattern, different contrast agents can be employed to image specific regions of the animal body. For example, conventional, FDA-approved, MRI contrast agents, such as gadolinium chelates, are extracellular agents that quickly distribute into the blood pool and interstitium following their intravenous injection. These agents do not cross the intact blood–brain barrier, and have found applications for imaging of MR angiography and enhancement of lesions such as tumors.⁷⁸ Blood-pool agents have a relatively large size and longer half-life in the blood than extracellular fluid (ECF) agents and include gadolinium complexes with albumin, polymers, etc.⁷⁸ A number of organ-specific contrast agents are also available. Gadolinium-, manganese (Mn²⁺) - and super-paramagnetic iron oxide (SPIO)-based agents are available for hepatobiliary imaging.^78,80 Ultra-small super-paramagnetic iron oxide (USPIO)- and gadolinium-based agents are available for lymph node and bone marrow imaging.⁷⁸ Iron oxide nanoparticles are taken up by the monocyte-macrophage system and can be used as MRI markers of inflammatory and degenerative disorders characterized by enhanced phagocytic activity.^79,80 Mn²⁺-based agents are particularly useful for neuroimaging.⁷⁸ Mn²⁺ can enter the central nervous system neurons and upon neuronal excitation, gets released into the synaptic cleft. This phenomenon is used to extract information on the axonal transport and trans-synaptic activities. ^81,82 (see Figure 3). Orally administered agents are available for gastrointestinal imaging.⁷⁸ Oxygen⁸³ and hyperpolarized gases such as ³He and ¹²⁹Xe have been employed for pulmonary ventilation studies.^84,85

FIGURE 3.

Axial magnetic resonance images of a normal adult male rat brain acquired before (left) and 24 h after (right) intraperitoneal injection of 45 mg/kg MnCl₂. MnCl₂ results in regions enhancement (arrows). Images were acquired with a conventional T1-weighted spin echo acquisition (TR/TE = 600/10 ms) at 7T.

A new generation of contrast agents consisting of an assembly of a contrastophore conjugate and a target-specific ligand are now being developed for molecular imaging.^79,86 A number of carrier molecules such as antibodies, peptides, polysaccharides, aptamers, and synthetic compounds can be used to deliver the magnetic labels to the specific targets.⁷⁹ Tagging of various gadolinium- and nanoparticles-based contrast agents have been performed to monitor the stem cells and immune cells in vivo.⁸⁷ Investigators have also been able to insert magnetic reporter genes into the cells resulting in expression of iron-storage proteins to be detected on MRI images.^88–90 A group of smart contrast agents, whose relaxivity varies depending upon various local factors such as temperature, pH, redox potential, enzymes, etc. have also been developed.^78,91 Copper- ⁹² and zinc-responsive ⁹³ contrast agents have also been developed.

Another MRI contrast mechanism, chemical exchange saturation transfer (CEST) originates from the chemical exchange of endogenous amide or hydroxyl protons or from exchangeable sites on exogenous CEST agents.⁹⁴ Paramagnetic CEST (PARACEST) and hyperpolarized CEST (hyperCEST) are variations of CEST in which saturation transfer arises from presaturation of exchangeable protons on exogenous paramagnetic or hyperpolarized gas molecules, respectively.⁹⁴

Applications

High-resolution (spatial resolution <100 μm) MRI, also called MR microscopy (MRM) has been performed in small animals (see Figure 4). A desired combination of a number of structural imaging (i.e., T1-, T2- and spin density-weighted, fat saturated and fluid attenuated, etc.), DTI, perfusion MRI, DCE-MRI, fMRI, and MRS of the region of interest can be performed in a single setting. A broad overview of applications of MRI for animal imaging can be found elsewhere.^80,95–99 MRM provides relatively high-resolution images of brain because of the absence of cardiorespiratory movements.⁹⁶ High-field MRM has recently been employed for murine brain imaging at a near single-cell layer resolution (30–40 μm) in vivo.⁶² DTI with high spatial and angular resolutions has been employed for fiber tractography in vivo.^100,101 fMRI has found applications to analyze neuronal activity in normal¹⁰² and diseased rats.¹⁰³ Mn²⁺-enhanced MRI has been employed for neural tract mapping, fMRI, and to investigate neuroarchitecture.⁸² Multiparametric MRI including perfusion MRI and diffusion-weighted imaging has been used to investigate small animal stroke models.¹⁰⁴ Perfusion MRI and CE-MRI can be employed to investigate small animal brain tumor models.^105,106

FIGURE 4.

High-resolution (500 × 100 × 100 μm) axial magnetic resonance imaging (MRI) image of dorsolateral prostate of transgenic mouse prone to prostate hyperplasia. Prostate region is outlined in green. The image was acquired by using rapid acquisition with refocused echoes (RARE) sequence (TR/TE = 5000/15 ms).

MRM has also been employed for cardiac,¹⁰⁷ embryonic,¹⁰⁸ and developmental¹⁰⁹ studies. Perfusion fixation can be employed to obtain regional or whole-body high-resolution MRM images.^{96 3}He-and ¹²⁹Xe-enhanced MRM is employed to investigate pulmonary ventilation patterns.⁹⁶ Morphologic phenotyping has also been performed by MRM.⁹⁶ Longitudinal MRM studies can be performed to analyze the temporal evaluation of various tumor parameters such as the total volume, vascular permeability, water diffusion and blood oxygenation, etc. Magnetic tagging has been employed to analyze the cellular migration in the brain,¹¹⁰ heart,¹¹¹ liver,¹¹² and kidneys¹¹³ of small animals in vivo. Multiparametric MRI including DTI and perfusion MRI has been used to characterize the rat model of reperfused partial liver infarction¹¹⁴ and liver tumors.^115,116 DCE-MRI is used to investigate the tumor microvasculature and can potentially be used as an imaging biomarker to measure anti-angiogenic effects of cancer treatments.⁶⁸ Affects of interventional procedures such as photodynamic therapy on tumors can also be studied using DCE-MRI (see Figure 5).

FIGURE 5.

Estimation of affect of photodynamic therapy (PDT) on U87 glioma implant in a rat model using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). DCE-MRI images of a U87 glioma implant in a rat model acquired before (left upper) and immediately after (left lower) intravenous injection of Gd-DTPA contrast agent. Analysis of the dynamic uptake (right) revealed that contrast uptake was increased as evidenced by higher peak signal level in MRI images acquired following PDT. The images were acquired using a fast low angle shot (FLASH) sequence (TR/TE = 200/3 ms).

Principle applications of MRS are for cancer therapy assessment and analysis of intrinsic metabolic pathways and drug metabolism.^76,99,117 Molecular quantification down to the millimolar range can be performed.^77,99 Because of its ability to detect molecules such as N-acetylaspartate, creatine, phosphocreatine, choline, lactate, inositols, alanine, glutamine, and glutamate, ¹ H MRS has found applications to investigate the molecular pathways in vivo.⁷⁷ Energy metabolism molecules such as phosphocreatine, adenosine tri-phosphate, phosphorylcholine, and inorganic phosphate (Pi) can be analyzed using ³¹P-MRS which makes it particularly useful to study tissue energetics.^{77 13}C-MRS has relatively lower intrinsic sensitivity and is used to study compounds such as glycogen that are present in a high concentration inside the body.^{77 13}C-labeling is employed to study metabolic pathways. ¹⁷O-MRS is used to investigate the oxygen metabolism. ¹⁹F-MRS is used to investigate ¹⁹F-containing drugs such as anesthetic agents, gene activity, pH, metal ion concentrations, oxygen tension, hypoxia, vascular flow, and vascular volume.^118,119 Very high resolution MRS has also recently been performed to investigate the cellular protein content in vivo.¹²⁰ Further details of various MRS applications can be found elsewhere.^76,99,117

Challenges and Future Prospects

A very high development and maintenance cost of MRI and NMR scanners restrict their availability; oftentimes in vivo animal imaging systems are found in advanced centers only. The desire to obtain very high resolution images using MRM is particularly hindered by a steep reduction in the signal-to-noise ratio with reduced voxel size. The sensitivity of MRI techniques for functional and metabolic measurements is also relatively low when compared to PET and SPECT. Very long image acquisition times for the techniques such as MRS decrease the throughput of the animal research. At last, intrinsic cardiorespiratory movements of the animals are particularly a problem for imaging of the mobile organs.

MRI has an edge over the other imaging techniques because of its ability to provide a multiplicity of information by employing different imaging sequences. This makes MRI a versatile tool to define anatomy and pathophysiology of an animal single-handedly. Recent advances such as availability of MRI scanners with very high magnetic field and gradient strengths, MRI–PET scanners, dedicated small animal imaging microcoils, increased sensitivity of hyperpolarized gas imaging, and faster imaging mechanisms have revolutionized its applications. Development of nanoparticles-based target-specific contrast agents and innovative contrast mechanisms such as CEST is expected to further take small animal MRI research into a new horizon.

NUCLEAR IMAGING MODALITIES

Two main nuclear imaging modalities namely SPECT and PET are typically used for animal imaging. Dedicated PET¹²¹ and SPECT¹²² imaging systems have been developed for small animals. For image acquisition by these modalities, a radionuclide is injected into the body. The radionuclide distributes in the body based on its similarity to other compounds needed by the cells or tissues (e.g., bony phosphates, glucose, water, red blood cells). A SPECT radionuclide undergoes radioactive decay and emits gamma rays of a particular energy. The gamma rays are captured by an external camera to acquire a number of 2D projections from multiple angles which are combined together to form a 3D SPECT image.^123,124 On the other hand, a PET radionuclide undergoes positive beta decay and emits a positron. The positron travels for a short distance (approximately, a few millimeters), annihilates with an electron to emit a pair of photons (energy 511 KeV) that travel nearly 180° apart and are detected by externally placed detectors. The photons coincidently detected by two opposing detectors define a line of response (LOR). The distribution pattern of the LORs is used to reconstruct a 3D image of the radioactivity distribution within the body.^123,124

Radionuclides and Radiotracers

Extensive information regarding various PET radionuclides can be found elsewhere.¹²⁵ Traditional ¹³N-, ¹¹C-, and ¹⁵O-based radionuclides have a very short half-life (i.e., seconds or a few minutes) and can be used at the production site only; their short half-life prevents feasible transport to the remote imaging locations. ¹⁸F is the most commonly used PET radionuclid because of its low positron energy, ease of production, and relatively long half-life (50–100 min).¹²⁶ Considerable effort has been made to develop nonstandard radionuclides with a long half-life to study time-consuming processes such as protein metabolism and cell proliferation.^{126,127 68}Ge and ⁸²Sr, have a relatively long half-life (270.8 and 25.3 days, respectively).¹²⁶ Two As isotopes with a long half-life, ⁷²As (26 h) and ⁷⁴As (17.8 days), are also available.^128,129 The radionuclides can be labeled to a ligand with affinity toward a particular target molecule inside the body for applications such as drug development and to study the pathophysiology of various diseases. Extensive information regarding the process of development and evaluation of a target-specific PET radionuclide can be found elsewhere.¹³⁰ 2-Deoxy-2-fluoro-D-glucose (FDG), a glucose analog, is a commonly used ligand.^130–133 Radiolabeled-FDG enters the cells through the glucose transporters and gets phosphorylated by hexokinase. In tissues with low glucose 6-phosphatase, the phosphorylated FDG becomes trapped within the cells. [¹⁸F]FDG is the most commonly used agent in PET imaging.¹³⁴ Its variant 6-deoxy-6-[¹⁸F]fluoro-D-glucose does not undergo intracellular phosphorylation and is used to image glucose transport dynamics¹³⁵ (see Figure 6). ⁸²RbCl is a potassium analog that is rapidly taken up by the muscle cells.¹³⁶ Na[¹⁸F]fluoride is used for bone imaging.¹³⁷ Radiolabeled bombesin and its derivatives have been explored for cancer PET imaging.¹³⁸ PET radionuclides based upon thymidine kinase pathways, matrix metalloproteases are used for PET imaging of angiogenesis.¹³⁹ Radiolabeled peptide conjugates for PET imaging of apoptosis¹⁴⁰ and tumor protein metabolism¹³⁸ have also been developed.

FIGURE 6.

Positron emission tomography (PET) images of a rat injected with 6-deoxy-6-[¹⁸F]fluoro-d-glucose ([¹⁸F]6FDG) and scanned during glucose clamp under fasting (left) and hyperinsulimic (right) conditions. The tissue concentration of [¹⁸F]6FDG was elevated in response to insulin stimulation (Images courtesy of Dr. Raymond Muzic, Case Western Reserve University, Cleveland, OH, USA).

Radiotracers based upon radioactive metals such as ^99mTc, ¹¹¹In, ¹⁸⁸Re, ¹³¹I, ¹²³I, ²⁰¹Th, and ¹³³Xe are often used for SPECT imaging. Specific radiotracers have been developed for SPECT imaging of individual receptors and transport systems in the brain.¹⁴¹ [^99mTc]-Sestamibi¹⁴² and [¹¹¹In]-pentetreotide¹⁴³ have been employed for myocardial perfusion and neuroendocrine tumor imaging, respectively. 2-Nitroimidazole and nitric oxide derivatives have been used to image tumor hypoxia and inflammation-related changes, respectively.¹⁴⁴ Radiolabeled bombesin and its derivatives have been explored for cancer SPECT imaging.¹³⁸ Radiolabeled peptide conjugates for SPECT imaging of apoptosis,¹⁴⁰ amyloid plaques,¹⁴⁵ and tumor protein metabolism¹³⁸ have been developed.

Applications

Nuclear imaging modalities can provide quantitative information regarding molecular concentrations by estimating the number of annihilation events or incident gamma photons in vivo. The sensitivity of a preclinical PET for the detection of annihilation events is 2–4% and provides images with a spatial resolution of 1–2 mm,^123,146 whereas the sensitivity of a preclinical SPECT is approximately 0.3% and provides images with a spatial resolution of ≈0.5 mm.^123,146 A broad comparative analysis of relative strengths and limitations of SPECT and PET can be found elsewhere.^146,147 PET is 10 times as sensitive as SPECT. On the other hand, availability of large number of radionuclides, simultaneous imaging of multiple agents, cost-effectiveness, and adaptability to specific imaging situations are amongst the few advantages of SPECT. Overall, PET is advantageous over SPECT for drug development and cancer imaging.

PET offers sensitivity in the picomolar range. It has been employed to study myocardial gene expression and effect of its phenotypic expression on the cardiac parameters.^148–150 Cardiac and respiratory gating has also been employed to improve image quality.¹⁵⁰ Neurological applications of PET include monitoring of neurotransmitter function, molecular events in brain tumors, and angiogenic and gene therapies, among others.^148,149 It is employed for tissue characterization, staging, and therapy control in oncology.^148,149,151 Radioarsenic-labeled bavituximab can be used for PET imaging of solid tumor vasculature.¹⁵² Similarly, applications of SPECT have also been described for drug development, and brain and cardiac imaging.^153,154 It has also been used for studies related to gene expression and cancer; however, PET is more preferred nowadays.¹⁵³ Bone imaging has also been performed using SPECT. Nuclear imaging modalities have also been employed for imaging of atherosclerotic lesions,¹⁵⁵ amyloid plaques of Alzheimer’s disease,¹⁵⁶ and reporter gene imaging.¹⁵⁷ A tumor can be differentiated from inflammation using various proliferation markers.¹⁵⁸ Tumor radiosensitivity can also by monitored by investigating hypoxia and apoptosis.¹⁵⁹

Challenges and Future Prospects

Although nuclear imaging modalities can be used to reveal molecular events, they are unable to provide any information regarding molecular structures. For characterization of molecular structures, other techniques such as high performance liquid chromatography (HPLC) need to be used. Further developments in detector technology and image reconstruction algorithms for nuclear imaging modalities are desired to overcome the limitations such as very low detection sensitivities and spatial resolutions. Development of highly target-specific ligands is desired to radiolabel the radionuclides. High costs for PET scanners and short half-life of radionuclides have limited small animal PET imaging to the advanced centers only. Very short half-life of PET radionuclides necessitates the presence of a cyclotron in vicinity of the PET imaging system. Pharmacologic constraints, and prerequisites and interspecies variations in metabolism and mass effects of PET ligands also remain a challenge for extension of results of the animal studies into the clinical practice.¹⁶⁰

On their positive aspect, nuclear imaging modalities provide a uniform assay system to evaluate pharmacokinetics and pharmacodynamics of the candidate drugs and molecular probes in animal models. Dynamic data acquisition enables complete description of kinetics of new agents in a single animal. Development of new radionuclides and target-specific ligands is likely to play a major role in the understanding of mechanisms of diseases such as myocardial infarction, Parkinson’s disease, and Alzheimer’s disease and their management.¹⁵⁷ Molecular information from these modalities can be combined with the structural and molecular information provided by CT and MRI. Image registration mechanisms, dual-modality imaging systems, and contrast agents are expected to give promising results in animal imaging as described in Section ‘New Frontiers in Multimodality Imaging.’

ULTRASOUND IMAGING

In US imaging, sound waves of a desired frequency (determined by factors including depth of the organ of interest and the desired spatial resolution) are applied from a transducer upon the animal body, a fraction of which gets reflected from the acoustic interfaces (created by difference in stiffness and density variations of the tissues). A part of the reflected waves is detected by the transducer. The intensity pattern of the reflected waves reaching the transducer is used to generate a US image.^161,162 Different US imaging modes such as amplitude and brightness modes can be employed to construct 1D and 2D US images, respectively.^163–165 A number of Doppler US techniques have also been developed in which frequency shift of the sound waves scattered from moving particles (such as blood) is used to determine the velocity of the moving particles.^162,165 Finally, high intensity focused US (HIFU) is a technique in which a high intensity of US is imposed upon the region of interest that results in thermal heating (up to 60°, which leads to coagulation necrosis) and mechanical effects including cavitation and microstreaming at the focal point.^166,167

Ultrasound Contrast Agents

US contrast agents (UCAs) enhance image contrast by increasing backscatter of the sound waves. Detailed description of physical interactions between the UCAs and various US imaging modes can be found elsewhere.^168,169 A UCA typically consists of a gas core surrounded by a biocompatible stabilizing material such as a protein, lipid, or biopolymer.^169,170 Unfortunately, because of their large size, the UCAs cannot cross from the vascular pool into the extravascular space and hence have applications for vascular pool imaging only. Nontargeted UCAs have been used as red cell tracers.¹⁷¹

Target-specific UCAs are developed by using various strategies such as surface property modifications or covalent linkage of targeting ligands.¹⁷² A number of inflammation-specific UCAs have been developed by targeting the leukocytes or inflammation-specific adhesion molecules.¹⁷³ Markers such as vascular endothelial growth factor (VEGF) receptors have been exploited to develop angiogenesis-specific UCAs.¹⁷³ Fibrin- ¹⁷⁴ and platelet-targeted ¹⁷⁵ UCAs have also been developed. Recently, stimulus-responsive UCAs have been developed to increase the blood half-life of a drug as well as for US targeting by incorporating a brush border to the lipid shell that protects the targeting ligands from opsonizing factors in the absence of sound waves. Once the UCA is bound to the target, it can be destroyed allowing for the drug payload to be delivered.¹⁷⁶ The UCAs are a powerful tool for delivery because not only can they be targeted and monitored, but their insonation parameters can also be adjusted to induce oscillations that rupture the UCA, releasing the drug/gene payload.^177–179

Sonoporation or cellular sonication is a technique that via application of the US in the presence of a UCA, transiently enhances local cell membrane permeability by creating pores in the cell membranes and microvasculature,¹⁷³ which makes it a promising tool for image-guided drug and gene delivery into the cytoplasm. Large DNA particles as well as nanospheres of diameters up to 75 nm could enter a cell upon sonoporation.¹⁸⁰

Applications

US using the sound waves of 20–40 MHz can provide images with a spatial resolution as high as 30 μm in small animals.¹⁷³ In small animal imaging, the depth of penetration limitations associated with high-frequency US are not as problematic as they would be in human imaging. Molecular imaging of cardiovascular diseases is an established application of US and can also be monitored with contrast-enhanced US. Other pathophysiologic processes that can be monitored with contrast-enhanced US are atherosclerosis, angiogenesis, ischemia, inflammation, and thrombosis.^171,172 For example, ICAM-1 and P-selectin targeted UCAs are used to evaluate transplant rejection and ischemia-induced chest pain, respectively. Atherogenesis-associated inflammatory changes can be exploited to characterize atheromatous plaques by US imaging. VCAM- and lipid-UCAs have been used to identify the atheromatous plaques in their early stages and ischemic-reperfusion injuries, respectively.¹⁷¹ Platelet- and fibrin-specific UCAs can be used to improve clot visualization.^174,175 Ultrasound-induced increase in tPA activity has been exploited for stroke lysis.¹⁸¹ Tumor growth and its response to therapy can be monitored using angiogenesis-specific UCAs. Several techniques have been developed to monitor tumor neovascularization using US imaging^182–184 (see Figure 7). US can also be used to differentiate tumor therapy-induced inflammatory changes from the residual tumor by using phosphotidyl serine (PS)-modified UCAs because of increased PS receptors at the inflammation site (see Figure 8). Contrast-enhanced US has also been employed to measure the regional blood flow.¹⁸⁵ Detailed information regarding oncologic applications of HIFU can be found elsewhere.¹⁶⁶ Pulsed-HIFU has been employed for targeted drug delivery.^186–188 Sonoporation provides a safe alternative to viral vectors for delivering the nuclear acid material into a cell.^177–179 Sonoporation has also been used to disrupt the blood–brain barrier for more efficient drug delivery to the central nervous system. The increased permeability lasts for approximately 4 h after sonication, and molecules up to 40 kDa can be transferred through the pores.^189–191

FIGURE 7.

Contrast-enhanced US images of microvasculature of a rat tumor at different delay times after the US contrast agent (UCA) injection. The microvasculature can not be seen in the image acquired at the beginning of the UCA injection (left), whereas, it can be properly delineated in the images acquired after delay times of 3 (center) and 15 s (left) (Images courtesy of Dr. Agata Exner, Case Western Reserve University, Cleveland, OH, USA).

FIGURE 8.

Contrast-enhanced US images of a rat tumor obtained 7 days after performing the ablation therapy. Inflammatory region can readily be identified because of US contrast agent (UCA) targeting by phosphatidyl serine. The images were taken 12 min after injection of phosphatidyl serine UCAs. (Images courtesy of Dr. Agata Exner, Case Western Reserve University, Cleveland, OH, USA).

Noncontrast-enhanced US imaging has also found a few applications for small animals. For example, US elastography provides useful information regarding mechanical properties of the tissues and can be used to investigate suspicious mass lesions.¹⁹² Intravascular US elastography has been evaluated for characterization of atherosclerotic plaques and mural thrombi in rabbits.¹⁹³ In vivo characterization of atherosclerotic plaques have also been performed using US.¹⁹⁴

Challenges and Future Prospects

Operator dependence and challenges to provide whole-body images are the principle pitfalls of US. UCA applications in drug and gene delivery are limited by their relatively low loading capacity. Various strategies such as the utilization of a polymer shell have been employed to improve their loading capacity.¹⁷⁹ Also, the UCAs are confined to the vascular compartment only and cannot be used for extravascular or intracellular imaging.¹⁷³ On the other hand, primary advantage of the UCAs is that the signal generated from contrast-enhanced US will be a true indicator of the vascular bed. Noninvasive targeted drug and gene delivery by US can help in formulating new strategies for administration of drugs that would otherwise have a high incidence of adverse effects. US molecular imaging also has an edge over nuclear imaging techniques because of its relatively high spatial resolution, requirement of very low contrast dose, and absence of ionizing radiation exposure.

OPTICAL IMAGING TECHNIQUES

A number of optical imaging techniques have been developed in which the fundamental concept behind image formation is the detection of a photon emitted from inside the animal body by an external camera. In BLI, the photon is emitted upon metabolization of a substrate by a bioluminescent reporter enzyme.^195,196 In FLI, a fluorophore gets excited by absorbing a photon, and releases another photon of comparatively higher wavelength while returning back to the ground state.¹⁹⁵ OCT is based upon the physical principle similar to that of brightness mode of US imaging except that it uses light photons instead of sound waves.¹⁹⁷

Optical Reporters

Firefly luciferase (Photinus pyralis; FLuc) (half-life ≈2 h) is the most widely used reporter enzyme for BLI.^198,199 It catalyzes the formation of a luciferin–ATP complex, which gets oxidized leading to production of a cyclic peroxide compound that eventually becomes high-energy oxyluciferin. Oxyluciferin emits light photons of a broad spectrum (peak λ_max ~ 560 nm).^198,199 The light emission peaks ≈10–12 min after the luciferin injection and gradually decreases over next 60 min.²⁰⁰ Various ATP-independent luciferases such as sea pansy Renilla reniformis luciferase (RLuc),²⁰¹ click beetle Pyrophorus plagiophthalamus, and marine copepod Gaussia princeps Luciferase (GLuc)^202,203 have also been cloned. As compared to the firefly luciferase, ATP-independent luciferases can be easily fused with other proteins. However, their limited distribution, rapid kinetics, and higher background noise are a few limitations.¹⁹⁹ Bacterial lux operons encode all proteins required for bioluminescence. Hence, lux operon-expressing bacteria do not require an exogenous substrate to produce light. This fact has been exploited to investigate bacterial pathogenesis; however, attempts to incorporate lux operons into mammalian cells have not been successful yet.¹⁹⁹

Both endogenous (e.g., hemoglobin or collagen) as well as exogenous molecules can act as flourophores for FLI.¹⁹⁹ An exogenous fluorophore molecule can be an organic dye,²⁰⁴ a fluorescent gene reporter,²⁰⁵ a fluorescent protein,²⁰⁶ or a quantum dot.²⁰⁷ Recently, smart probes, whose conformation changes upon binding to a target (which increases their fluorescence intensity), have also been developed.^208,209 Fluorescent probes emitting light in the near infrared region (NIR) are now being produced to improve the depth of in vivo FLI.^210,211 A number of green fluorescent proteins (GFPs) have been developed, which are useful in studying gene expression and protein interactions.^212–215 In addition, red fluorescent proteins (RFPs) have also been developed.²¹⁵ The fluorescent proteins can undergo various changes such as photoactivation, photoconversion, and photoswitching.²¹⁵ Various mechanisms have also been described for contrast enhancement in OCT images.^197,216

Applications

Optical imaging techniques can detect optical reporters in concentrations as low as picomolar to femtomolar ranges.¹⁹⁹ Spatial localization of the optical signal can be performed by using other imaging modalities such as planar X-ray imaging. A broad overview of the applications of BLI for small animal imaging can be found elsewhere.¹⁹⁶ Luciferase complementation can be used to monitor protein interactions and their disruption during intracellular signaling or in response to therapy.^199,217,218 Interactions between drugs and their respective receptors can be studied using luciferase-based novel sensors.^219,220 BLI can be used to study the proteosome function and its response to therapeutic drugs,^221,222 gene expression, ^223,224 and for T cell imaging ²²⁵ in vivo. Dynamic BLI has been employed for tumor drug development to study their effect on the neoplastic cells.²²⁶

NIR fluorescent probes have been used to detect protease activity in vivo and are powerful tools for spatial and temporal analyses of molecular mechanisms of disease activity.^227,228 GFP-FLI can be applied for characterization of spatial and temporal dynamics of gene expression, protein localization, and protein–protein interactions in living cells.^212,213 Fluorescence resonance energy transfer (FRET) is an established technique to study protein–protein interactions.²¹⁵ Fluorescence signals from tumors can be quantified (see Figure 9). FLI has also been used to track tumor growth in real time^229,230 to analyze tumor–host interactions and to study the tumor response to therapeutic agents.²³¹ GFPs and RFPs are particularly useful for labeling of primary and metastatic tumor cells.²³¹ FLI has several advantages over BLI for tumor imaging.²³¹ Intravital microscopy can be performed at the single cell level^231,232 and can be used for real-time tracking of tumor angiogenesis in vivo.²³¹ Multiphoton laser scanning microscopy can also be performed to detect tumor angiogenesis in vivo through a window created on the skin surface.²³³ In vivo stem cell tracking has also been performed using FLI.²³⁴

FIGURE 9.

Spectral analysis can be performed to quantify tumor-specific probe activation using a Maestro^™ small animal in vivo imaging system. Here, a quenched activity-based imaging probe was injected intravenously and labeled a tumor in vivo. Total fluorescent signal from the tumor and the normal skin is quantified. (a) Spectral curves for each fluorophore as well as for autofluorescence can be generated and saved into a spectral library. Individual curves: white, mouse body; black, imaging stage; red, probe in vivo; magenta, pure probe. (b) By using spectral decomposition and artificial coloration for each spectral curve, a composite image is obtained. A ‘hotmap’ of the probe in vivo shows total fluorescent intensity of the imaging probe within the tumor. (c) Regions of interest can be delineated manually for quantitative analysis of images. (Image courtesy of Jennifer Cutter, NFCR Center for Molecular Imaging, Case Western Reserve University, Cleveland, OH, USA).

OCT has been applied for gastrointestinal, ocular, and skin imaging. However, because of very low depth penetration, its applications are limited to superficial imaging only.^197,235

Challenges and Future Prospects

Limited depth penetration, restricted image planes, and nonlinear relationships between signal intensity, tissue heterogeneity, and depth are a few limitations of optical imaging techniques. Although these techniques enable detection of molecules in concentrations within femtomolar ranges, the information provided by these techniques is qualitative. The use of GFPs can confound the experimental results by altering the transcription and translation rates of the cells.²³⁶ In addition, the surrounding protein cavity (steric hindrance) can also influence the fluorescence of the fluorophores.²³⁷ The major advantages of optical imaging techniques are their relatively low operational costs, simple imaging procedures, absence of ionization radiation exposure, quick image acquisition, and ability to image multiple animals repetitively over a long period of time. Recent development of fluorophores in the NIR range has allowed for imaging of tissues as deep as 7–14 cm.²³⁸ Intravital microscopy can also be used to image deeper tissues in vivo.²³¹ Multispectral imaging has enabled separation and quantification of multiple fluorophores at the same time.²³⁹ Recent developments in BLI concentrate primarily on creating red-shifted luciferase reporter systems for more efficient light transmission through the tissues. Techniques for real-time BLI of moving rodents²⁴⁰ and dynamic BLI^226,241 have also been developed in the recent past. Today, optical imaging is transforming the way investigators translate gene and protein expression into real-time events.

NEW FRONTIERS IN MULTIMODALITY IMAGING

In order to synergize the information provided by different kinds of images, images produced by two or more techniques are often combined together. There are two common options for combining the images. The first option is coregistration, or fusion, of images acquired from different imaging systems (e.g., MRI and PET, SPECT, and CT).^242,243 This works well for rigid and relatively immobile organs such as the brain. However, for the thorax, abdomen, and pelvis, where organs are relatively more mobile and deformable, the process of registration of sequentially acquired images becomes more complicated. This leads to the second option: development of new imaging systems for simultaneous or sequential imaging of the same volume with minimal movement.²⁴³ Simultaneous imaging is also important for the analysis of dynamic processes such as contrast enhancement, where sets of information need to be acquired simultaneously at different point of times. Various dedicated dual-modality imaging systems have been designed for small animal imaging. In dedicated SPECT–CT¹⁵⁴ and PET–CT²⁴⁴ (see Figure 10) systems, structural data from CT images are also used to generate attenuation correction maps for SPECT and PET image reconstruction, respectively. Dedicated SPECT–CT systems have found applications for oncology research,¹⁵⁴ development of radiolabeled nanobodies²⁴⁵ and gene therapy vectors,²⁴⁶ and investigation of various disease models.²⁴⁷ PET–CT is principally employed for oncologic imaging.²⁴⁸ MRI compatible optical imaging systems have been developed.²⁴⁹ Dedicated PET–MRI,²⁵⁰ SPECT–MRI,²⁵¹ and optical-PET²⁵² systems have also been developed for small animal imaging. PET–MRI imaging would be particularly useful for animal research. It provides an opportunity to combine functional information from PET with anatomical as well as various other kinds of information (i.e., functional, metabolic, or perfusion) from MRI.²⁵³ Multimodality contrast agents are also being developed nowadays which enable analyses of the same event by multiple imaging techniques. A number of dual-modality probes have been described for in vivo molecular imaging.^254,255 Multimodal contrast agents have been developed for corroboration of MRI with other modalities.^15,256 A novel triple-modality reporter gene for whole-body fluorescent, bioluminescent, and nuclear imaging have been reported.²⁵⁷ Synergistically integrated nanoparticles are being projected as an excellent candidate to be multimodal probes for nanobiotechnology.²⁵⁸

FIGURE 10.

Simultaneous in vivo 18F-FDG-positron emission tomography (PET) and computed tomography (CT) scans of a mouse 50 min after injection of 8.5 MBq of [¹⁸F]FDG. Note the strong signal in the kidneys and bladder in the PET images (transverse slice in (b), coronal slice in (e)) and the corresponding position of the kidneys and bladder in the CT images (transverse slice in (a), coronal slice in (d)). The fused transverse slice is shown in (c) and the coronal slice in (f). The scan was acquired using 200 views within a total scan time of 18 min. (Reprinted with permission from Ref 259. Copyright 2002 IOP Publishing Ltd.).

CONCLUSION

Comparative analyses of the basic principle of image formation, image acquisition times, important applications, strengths, and limitations of the individual imaging techniques can be found elsewhere.^1,2 Maximum spatial resolution, sensitivity, and tissue depth penetration of these techniques have been compared in another review.²⁵⁸ None of the imaging techniques are free of limitations or serves every purpose of image acquisition from all areas of the animal. PET and SPECT have very high sensitivity, down to picomolar and nanomolar concentrations, respectively. Superb sensitivity enables application of these imaging techniques for quantitative analysis of tumor cell metabolism. Availability of a variety of probes and strategies confer PET a high degree of versatility. The ability to detect multiple probes simultaneously and the availability of radioisotopes with longer half-life are some of advantages of SPECT over PET; however, they are limited in their spatial resolution.

Amongst structural imaging techniques, MRI provides images with high spatial resolution and can provide anatomical, functional, metabolic as well as molecular information in the same setting. Exposure to RF waves and magnetic fields does not affect animal pathophysiology to a significant extent. μCT also provides images with excellent spatial resolution and is particularly useful for imaging of bones and lungs. Although not typically a significant issue in clinical CT imaging, very high X-ray doses required for μCT imaging could decrease the life expectancy and confound the experimental results by altering the pathophysiology of the animals. Ultrasound is more applicable to analyze tissue physiology relatively close to the surface of the live animal in real time. Optical imaging techniques provide molecular sensitivity down to femtomolar ranges. BLI provides a relative measure of cell viability or cell function and has a high throughput. FLI is a versatile method that can perform multiplexed imaging by using multiple reporter wavelengths.^1,2 Tomographic (i.e., image slices) techniques are only now emerging with BLI and FLI. However, low tissue depth penetration and qualitative nature of results are significant drawbacks of optical imaging techniques.

In conclusion, we provided a broad overview of various imaging techniques and their emerging applications for whole animal research. Information provided here is expected to provide an insight into the vast field of whole animal imaging to scientists involved in translational research. Translational research plays a vital role in translation of scientific research into practical applications (bench-to-bedside) and hence enables development of new techniques for management of human ailments. Overall, in vivo biomedical imaging techniques provide an important tool in the armamentarium of biomedical researchers today, and offer great potential for future impact on advancement of human society.