Magnetic resonance molecular imaging
| Magnetic resonance molecular imaging |
Editor in Chief: Omar Zurkiya, M.D., Ph.D. [1]
Introduction
Magnetic resonance imaging (MRI) is regularly used to construct an image based on the intrinsic contrast provided from the relaxation of spin of hydrogen atoms. These images provide an accurate anatomical picture that has greatly advanced health care today. Clinically, the images produced by MRI are invaluable, but the full potential of using MRI in acquiring functional, physiological, and molecular information is only beginning to be realized. The goal of this work is to further understand and develop approaches to magnetic resonance imaging to acquire information on the molecular level, a field termed molecular imaging.
Molecular imaging refers to the study of cellular and molecular events through noninvasive investigation [1]. Bringing molecular imaging to MRI requires generating contrast specific to information on the molecular level that is also sensitive enough to be observed at the voxel level. Gadolinium ion-based contrast agents are required to be present in relatively high concentrations, on the order of millimoles. Targeted in vivo imaging with MRI, however, often requires contrast agent compounds to be effective at concentrations on the order of 10 μM [2]. It is therefore desirable to obtain contrast agents in which a small number of bindings to targets will be sufficient to generate contrast. Iron oxide nanoparticles provide the possibility of detection at the single particle level [3]. These particles, however, are at the micrometer size. Such large particles are likely to interfere with normal cellular processes limiting their application to cellular and molecular imaging. It is therefore important to devise methods for imaging small particles, on the order of nanometers in size. This is especially important for intracellular studies.
Here we will begin by looking at contrast in MRI. The parameters described that are responsible for generating the intrinsic contrast between tissues in MRI are the same parameters affected by synthetic contrast agents. This will help to understand the mechanism of contrast agents and the reason superparamagnetic iron oxide (SPIO) nanoparticles are an appropriate choice for molecular imaging.
Contrast in MRI
The principles of MRI follow from the same principle of exciting spins as found in nuclear magnetic resonance (NMR) [4][5]. In most cases, the MRI machine is tuned to image the most abundant spin population found in the body, that of the protons of bulk water. Contrast in MRI arises from the difference in signal in adjacent pixels or voxels due to three basic parameters, spin density, T1, and T2 (Figure 1.1). The spin density, proportional to M0, reflects the total number of spins available to be imaged within a given tissue (or pixel). The system of spins is excited by radiofrequency (RF) energy, followed by a period of spatial encoding and then a period of signal acquisition during which the spin system is emitting energy as it relaxes to its equilibrium state. This energy is received as the signal for the MRI image.
The signal therefore depends not only on the spin density, but also upon two decay parameters, T1 and T2, which describe the rates at which the spin system is changing. T1 is rate at which the system returns to equilibrium following excitation by RF energy. The faster the system returns to equilibrium, the more spins available to be excited by the next imaging pulse. The effect of T1 on the signal is indicated in Figure 1.1b. For a given amount of time, a pixel with a shorter T1 would result in greater signal.
T2, often called spin-spin relaxation, refers to decay time of the signal received. During signal acquisition, the effect of field inhomogeneity, mostly arising from nearby spins affecting each other, causes the coherence of the signal to be lost. This decay rate is indicated in Figure 1.1c where, for a given amount of time, a pixel with a shorter T2 would result in less signal compared to an identical pixel in which all other parameters were the same.
Putting these parameters together results in the most basic signal equation:
<math>S\;\;\propto{}\;\;M_{o}\;(1-\exp{\frac{-TR}{T_{1}}})\;\exp{\frac{-TE}{T_{2}}}</math> (1.1)
Here, TR is the repetition time and TE is the echo time of the imaging pulse sequence. This equation represents the effects described above. The signal (S) is proportional to spin density (M0). It is increased by a shorter T1 and decreased by a shorter T2. By choosing TR and TE carefully, images can be chosen to favor T1, T2 or spin density weighting. Table 1.1 shows the decay rates of several tissues at 1.5T and 37οC. The well-known utility of MRI in imaging soft tissue arises from the natural differences in decay rates of various tissues, as in the case of brain imaging where gray and white matter can be easily distinguished. Figure 1.2 shows some examples of the various weighting of typical MRI images.
Table 1.1: Typical decay rates for tissues at 1.5T and 37οC. R1≡1/T1, R2≡1/T2. (data from [5])
| Tissue | T1 (ms) | R1 (s-1) | T2 (ms) | R2 (s-1) |
| gray matter | 950 | 1.05 | 100 | 10 |
| white matter | 600 | 1.67 | 80 | 12.5 |
| Muscle | 900 | 1.11 | 50 | 20 |
| cerebrospinal fluid (CSF) | 4500 | 0.22 | 2200 | 0.45 |
| Fat | 250 | 4.00 | 60 | 16.67 |
| Blood | 1200 | 0.83 | 100-200 | 5-10 |
Since these are the basic parameters of contrast in the MR image, it is these parameters that should be altered by any potential contrast agent. The most common MR contrast agents are paramagnetic gadolinium ion complexes and superparamagnetic iron oxide particles (see Table 1.2). Gadolinium contrast agents are used primarily for their ability to shorten T1, although at high concentrations, they can have significant effects on T2 as well. Iron oxide agents, on the other hand, have less effect on T1 and are used primarily for their effect on T2. Based on equation 1.1, it can be seen that gadolinium enhanced T1 imaging would show bright pixels in locations where contrast is present, and iron oxide enhanced T2 imaging would show decreased signal where contrast agent is present. In either case, it is the contrast with adjacent pixels that is the desirable effect.
In practice today, these contrast agents are non-specific. They are taken up into tissues based on passive biodistribution. Oral agents are useful for picturing bowel, and organs that filter the blood such as the liver and kidneys, have high concentrations of contrast when agents are injected intravenously. In order to extend the use of contrast agents to other, specific tissues, it is necessary to develop targeted probes. Such probes have the advantage of targeting and delineating areas of specific molecular activity on an MR image. For instance, in clinical procedures, an anatomical image revealing a mass would be routinely biopsied. If, however, there were a way to deliver a contrast agent that is only activated under specific, controlled circumstances, such as the expression of a carcinogenic genotype, one could obtain an image that provides molecular, as well as anatomical information. The need for biopsy would be reduced. This example represents a great advancement in clinical diagnosis, yet it represents a relatively modest goal of the potential such research could provide.
Table 1.2: Relaxivity values of common contrast agents. Gadolinium-based agents are generally used for their ability to shorten T1, while iron oxide based agents are generally used for their effect on T2. (data from [6])
| MR Contrast Agent | Main use | Molecular weight or particle size | Relaxivity (mM s)-1 | Target |
| Gd-DTPA | T1-agent | 0.6 kDa | r1= 3.7 | Extracellular |
| Dextran-Gd-DTPA | T1-agent | 75 kDa | r1= 11 | Blood-pool |
| Carboxydextran-coated SPIO SHU-555 | T2-agent | 62 nm | r1= 12, r2= 188 (0.94T) | Capillary permeability |
| Dextran-coated SPIO AMI-25 | T2-agent | 58 nm | r1=24, r2= 107 (0.47T) | MPS organs (liver) |
| Dextran-coated USPIO MION-46L | T2-agent | 18-24 nm (CLIO 30-40 nm) | r1= 16, r2= 35 (0.47T) | MPS organs |
| Dextran-coated USPIO AMI-227 | T2-agent | 17-20 nm | r1=23, r2= 53 (0.47T) | Lymph nodes |
| MION-encapsulated liposomes | T2-agent | 170-300 nm | r1= 23, r2= 130 (0.47T) | MPS organs (liver) |
| PEGylated magnetoliposomes | T2-agent | 40 nm | r1= 3, r2= 240 (1.5T) | Bone marrow |
| (Protein-coated) magnetoferritin | T2-agent | 12 nm | r1= 8, r2= 218 (1.5T) | Blood-pool |
Targeted imaging applications raise the need for development of new contrast agents capable of being detected at low concentrations. Successful in vivo imaging with MRI often requires target concentrations on the order of 10 μM [2]. In these applications, the contrast agent molecule is conjugated to the ligand which will bind the target, meaning the contrast agents must be effective at micromolar concentrations. The ability to use gadolinium for molecular imaging applications is therefore limited by its need to be present in millimolar quantities in order to cause a visible contrast. In Table 1.2, it can be seen that the r1 of gadolinium agents is on the order of tens of mM-1 s-1 (3.7—11 mM-1 s-1), whereas for superparamagnetic iron oxide nanoparticles (SPIO), the r2 is larger, on the order one to two hundred mM-1 s-1. The effect of SPIO nanoparticles is further magnified by the fact that these values are cited for mM Fe, but each nanoparticle contains several thousand iron atoms. Therefore, the relaxivity per particle is amplified by this factor, allowing SPIO to be an effective contrast agent even when molecular targets are present at low, micromolar concentrations.
Molecular Imaging
Molecular imaging, in general, refers to the study of cellular and molecular events through noninvasive investigation [1]. In MRI, molecular imaging is dependent on induced changes in proton relaxivity on the molecular and cellular level. Bringing molecular imaging to human MRI requires generating contrast on the molecular level that is not only specific, but also sensitive enough to be observed at the voxel level. MRI scanners at 3 Tesla, the current state of the art in human clinical scanning, offer submillimeter resolution. Although this resolution provides for detailed anatomical pictures, it presents challenges for molecular imaging. Probes must be able to generate contrast by specific molecular mechanisms, yet the change in signal must be large enough to be visible on the voxel level.
Over the last decade, biocompatible iron oxide particles have been linked to specific ligands for targeted molecular imaging applications [7][8][9][10][11]. However, due to their relatively large size and clearance by the reticuloendothelial system (RES), there is still a lack of widespread biomedical molecular application. Imaging of macrophage activity remains the most significant application, particularly for tumor staging of the liver and lymph nodes, and several commercial products are either approved or in clinical trials. Labeling non-phagocytic cells in culture using modified particles, followed by transplantation or transfusion in living organisms, has led to an active research interest to monitor cellular biodistribution in vivo, including cell migration and trafficking. While most of these studies have been more focused on establishing techniques, further use of these approaches will be as tools to obtain deeper insights into the dynamics of in vivo cell biology and at monitoring therapies that are based on the use of stem cells and progenitors.
References
- ↑ Weissleder, R. and U. Mahmood, Molecular imaging. Radiology, 2001. 219(2): p. 316-33.
- ↑ Blankenberg, F.G. and H.W. Strauss, Nuclear medicine applications in molecular imaging. J Magn Reson Imaging, 2002. 16(4): p. 352-61.
- ↑ Shapiro, E.M., S. Skrtic, K. Sharer, J.M. Hill, C.E. Dunbar, and A.P. Koretsky, MRI detection of single particles for cellular imaging. Proc Natl Acad Sci U S A, 2004. 101(30): p. 10901-6.
- ↑ Slichter, C.P., Principles of Magnetic Resonance. 1980, Berlin: Springer.
- ↑ Haacke, E.M., R.W. Brown, M.R. Thompson, and R. Venkatesan, Magnetic Resonance Imaging: Physical Principles and Sequence Design. 1999, New York: John Wiley and Sons.
- ↑ Mornet, S., S. Vasseur, F. Grasset, and E. Duguet, Magnetic nanoparticle design for medical diagnosis and therapy. Journal of Materials Chemistry, 2004. 14: p. 2161-2175.
- ↑ Weissleder, R., Target-Specific Superparamagnetic MR Contrast Agents. Magnetic Resonance in Medicine, 1991. 22: p. 209-212.
- ↑ Remsen, L.G., C.I. McCormick, S. Roman-Goldstein, G. Nilaver, R. Weissleder, A. Bogdanov, K.E. Hellstrom, I. Hellstrom, R.A. Kroll, and E.A. Neuwelt, MR of Carcinoma-Specific Monoclonal Antibody Conjugated to Monocrystalline Iron Oxide Nanoparticles: The Potential for Noninvasive Diagnosis. American Journal of Neuroradiology, 1996. 17: p. 411-418.
- ↑ Moore, A., J.P. Basilion, E.A. Chiocca, and R. Weissleder, Measuring transferrin receptor gene expression by NMR imaging. Biochim Biophys Acta, 1998. 1402(3): p. 239-49.
- ↑ Bulte, J.W., S. Zhang, P. van Gelderen, V. Herynek, E.K. Jordan, I.D. Duncan, and J.A. Frank, Neurotransplantation of magnetically labeled oligodendrocyte progenitors: magnetic resonance tracking of cell migration and myelination. Proc Natl Acad Sci U S A, 1999. 96(26): p. 15256-61.
- ↑ Artemov, D., N. Mori, B. Okollie, and Z.M. Bhujwalla, MR molecular imaging of the Her-2/neu receptor in breast cancer cells using targeted iron oxide nanoparticles. Magn Reson Med, 2003. 49(3): p. 403-8.
Acknowledgements
This page is first contributed by Omar Zurkiya, M.D., Ph.D.