MR Contrast Agents for Molecular and Cellular Imaging

Abstract

Although the principles of NMR were elucidated in the 1940s, the ability to produce images of humans necessitated the construction of large bore magnets in the late 1970s. Since the installation of the first MRI unit in the US, the hardware and software has continued to evolve, yet all the images produced by MRI depend on the same basic principles. A full discussion of the basic principles of magnetic resonance imaging is beyond the scope of this chapter; a short overview is provided here. In brief, protons (positively charged hydrogen nuclei) can be induced to emit signals, and this can be subsequently processed into images. MR takes advantage of the natural abundance of hydrogen atoms in (intra- and extra-)cellular water, lipids, proteins, and other more complex molecules, all of which are readily found in most tissues. When a subject is placed in a strong magnetic field the water hydrogen nuclei (protons) in the body align with the main (z) axis of the magnet in an equilibrium state. The protons spin and precess around the main axis of the magnetic field resulting in a net magnetization vector of which the frequency of precession is proportional to the strength of the magnetic field. To perturb the hydrogen nuclei from their equilibrium position, excitation radiofrequency (RF) pulses at the appropriate precessional (i.e. Larmor) frequency are used. The RF pulse originates from the head or body coil of the MR unit which acts as antennae to transmit pulses and receive signals from the body. This excitation effectively tips the proton spins away from the direction of the main magnetic field into a perpendicular plane or xy axis. The duration and magnitude of the RF pulse determines the degree of excitation produced. A 90 degree RF pulse brings the proton spins into the xy plane; a 180 degree RF pulse produces twice the degree of rotation (either into the z or opposite xy plane). The NMR signal is then induced in the coils surrounding the subject. These signals decay exponentially with specific time constants or relaxation times known as the longitudinal relaxation time Tl and the transverse relaxation time T2. The Tl and T2 relaxation times reflect the two distinct ways the NMR signal disappears following an RF excitation pulse returning the magnetization vector to its equilibrium position. By varying the pattern and timing of the RF pulse in combination with altering the main magnetic field using additional magnetic field gradients, one can impart spatial information to the NMR signals and thus create an MR image.