Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Acc Chem Res. Author manuscript; available in PMC 2010 July 21.
Published in final edited form as:
PMCID: PMC2713805

Gd-HOPO based high relaxivity MRI contrast agents


An external file that holds a picture, illustration, etc.
Object name is nihms-123174-f0001.jpg

Magnetic resonance imaging (MRI) is a particularly effective tool in medicine because of its high depth penetration (1 mm to 1 m) and ability to resolve different soft tissues. The MRI signal is generated by the relaxation of in vivo water molecule protons. MRI images can be improved by administering paramagnetic agents, which increase the relaxation rates of nearby water protons, thereby enhancing the MRI signal. The lanthanide cation Gd3+ is generally used because of its favorable electronic properties; high toxicity, however, necessitates strongly coordinating ligands to keep Gd3+ completely bound while in the patient. In this Account, we give a coordination chemistry overview of contrast agents (CAs) based on Gd-hydroxypyridinone (HOPO), which show improved MRI contrast and high thermodynamic stabilities.

Tris-bidentate HOPO-based ligands developed in our laboratory were designed to complement the coordination preferences of Gd3+, especially its oxophilicity. The HOPO ligands provide a hexadentate coordination environment for Gd3+ in which all the donor atoms are oxygen. Because Gd3+ favors eight or nine coordination, this design provides two to three open sites for inner-sphere water molecules. These water molecules rapidly exchange with bulk solution, hence affecting the relaxation rates of bulk water molecules. The parameters affecting the efficiency of these contrast agents have been tuned to improve contrast while still maintaining a high thermodynamic stability for Gd3+ binding. The Gd-HOPO-based contrast agents surpass current commercially available agents because of a higher number of inner-sphere water molecules, rapid exchange of inner-sphere water molecules via an associative mechanism, and a long electronic relaxation time. The contrast enhancement provided by these agents is at least twice that of commercial contrast agents, which are based on polyaminocarboxylate ligands.

Advances in MRI technology have made significant contributions to the improvement of clinical diagnostics by allowing visualization of underlying pathology. However, understanding the mechanism of a disease at the molecular level requires improved imaging sensitivity. The ultimate goal is to visually distinguish between different disease targets or markers—such as enzymes, hormones, proteins, or small molecules—at biologically relevant concentrations (millimolar to nanomolar). Although MRI techniques can provide images of the organs and tissues in which these biomarkers are regulated, the high sensitivity required to visualize the biological targets within the tissues is currently lacking—contrast enhancements of 50-fold beyond current agents are required to achieve this goal. According to the theory of paramagnetic relaxation, the contrast enhancement can be further improved by slowing the tumbling rate of the MRI agent. Theoretically, this enhancement would be greater for contrast agents with an optimal rate of water exchange. The Gd-HOPO-based contrast agents have optimal water-exchange rates, whereas the commercial agents have slower non-optimal water-exchange rates; thus, the Gd-HOPO agents are ideal for attachment to macromolecules, which will slow down the tumbling rate and increase contrast. This strategy has been recently tested with the Gd-HOPO agents via covalent attachment to virus capsids, affording contrast enhancements 10-fold beyond commercial agents.


Magnetic resonance imaging (MRI) is a frequently used diagnostic technique because it is non-invasive, able to differentiate between soft tissues, and has high lateral and depth resolutions.1-3 Despite these advantages the low sensitivity of MRI (mM) limits its use for targeted or molecular imaging.4 The signal intensity in MRI depends on the relaxation rates of water molecule protons. Differences in the environments of water molecules in vivo gives rise to the inherent contrast in MRI. This inherent contrast can be improved by the administration of paramagnetic metal ion containing contrast agents, which affect the relaxation rates of nearby water molecules. Among the different paramagnetic metal ions (Mn2+, Mn3+, Fe3+, and Gd3+) used in contrast agents, Gd3+ is most frequently used in commercial agents,5 because of its seven unpaired electrons and long electronic relaxation time. Though Gd3+ provides excellent contrast, the high toxicity of the Gd3+ aqua ion necessitates the strong coordination of Gd3+ by organic chelators before in vivo use.6 The contrast enhancing capacity of Gd3+-based contrast agents is directly proportional to the number of exchangeable water molecules in the inner coordination sphere of the Gd3+ ion. Unfortunately, the complexation of Gd3+ by organic chelators reduces the number of inner sphere water molecules. Hence, one of the major challenges in the field of Gd-based MRI contrast agents is to improve the contrast to obtain micromolar to nanomolar sensitivities for molecular imaging while minimizing the toxic side effects of Gd3+. Our work has focused on two aspects of this problem: 1) improving the stability of the Gd-complexes to be used as contrast agents by developing ligands based on the coordination properties of Gd3+; 2) maximizing contrast by tuning the parameters affecting signal intensity in MRI.

The measure of signal enhancement due to a MRI contrast agent is described by the term relaxivity, defined as the increase in the relaxation rate of water molecule protons per millimolar concentration of contrast agent applied. Gd-based contrast agents increase both the longitudinal (1/T1) and transverse relaxation rates (1/T2) of nearby water molecules. A greater increase is observed for the 1/T1 relaxation rates in tissues, hence Gd-based agents improve the contrast more in T1-weighted images than in T2-weighted images.5 The relaxation of the inner-sphere water molecules through dipolar interactions contributes the most to the longitudinal relaxation rate increase for Gd3+. According to the Solomon-Bloembergen-Morgan (SBM) equations which model paramagnetic relaxivity, the dipolar contribution to the longitudinal relaxivity (r1p) can be maximized by: increasing the number of inner sphere water molecules (q); optimizing the water exchange rates (1/τM, where τM is the water residence lifetime) between the inner sphere and outer sphere water molecules; and slowing the molecular tumbling rate (1/τR, where τR is the rotational correlation time) of the CA.7-10 These equations also describe the field strength dependence of relaxivity.

Relaxivity values up to 350 mM−1s−1 are theoretically possible, at field strengths between 20-100 MHz, when q = 3 and the electronic relaxation parameters are T1e = 15 ns and T2e = 0.3 ns. These high relaxivity values will help achieve the sensitivity required for targeted imaging using MRI. Relaxivity values for current contrast agents are of the order of 4-5 mM−1s−1 (20 MHz)5 which can be improved by tuning the parameters affecting relaxivity. Several attempts have been made at improving the commercial contrast agents, and these results are summarized in recent reviews.4, 11-14 Along with high relaxivity values, targeted MRI also requires the design of ‘turn-on’ or ‘responsive’ sensors that can afford a signal intensity change upon interaction with a biomarker of interest. Contrast agents that are responsive to pH, enzyme activity, and metal ions have been developed and the reader is referred to several recent publications on these topics. 15-19

Gd-HOPO-based contrast agents have been developed using a rational design based on Gd-coordination chemistry and these compounds will be the focus of this account. The mechanism of fast water-exchange observed for these complexes will be reviewed and its implications in obtaining high relaxivity values will be discussed. Recently published work involving the attachment of these complexes to biomacromolecules to attain high relaxivity values, and insights obtained into the dependence of relaxivity on molecular tumbling will also be reviewed. We will conclude with our perspective on strategies for attaining high sensitivity in MRI.

Coordination geometry considerations in designing ligands for Gd3+

O- versus N-donors

Trivalent lanthanide ions like Gd3+ usually exhibit variable coordination numbers of eight or nine.20 Lanthanide ions earlier in the periodic table have ionic radii similar to Ca2+ while the later lanthanides have ionic radii similar to Mn2+. Due to this similarity lanthanide ions can displace Ca2+ and Mn2+ from their binding sites in proteins, and are used to elucidate the structures of proteins by X-ray crystallography and NMR spectroscopy.21, 22 In a study by Pidcock and Moore, X-ray structures of Ca2+-and lanthanide-ion-containing proteins and coordination complexes were compared to find the similarities between the structural features of the metal binding sites.21 The average coordination number for Ca2+ in a protein is 6.0 (remaining ligands could be water molecules), while the average coordination number of a Ln3+ ion in an intrinsic protein Ca2+ site was found to be ca. 7.2. Oxygen atoms bind more strongly than nitrogen atoms in the coordination sites, with carboxylates (aspartates and glutamates) being the most common. A comparison of two crystal structures of a mannose binding protein is shown in Figure 1, depicting the Ca2+ sites in the native protein. Yb3+ binds to the oxygen donor rich intrinsic Ca2+ binding sites.23, 24 This similarity in the binding environments between Ca2+ and lanthanides was also used in a recent study, in which the Ca2+ binding sites in CCMV virus were filled with Gd3+ to create macromolecular MRI contrast agents.25 Thus in designing ligands for Gd3+ the structural features of natural Ca2+ binding sites can be incorporated and should lead to higher stabilities.

Figure 1
Learning from nature: Comparison of active sites in a mannose binding protein that recognizes high mannose concentrations on pathogenic cell surfaces. Left: Native rat mannose binding protein A (resolution 1.95 Å)24 showing a close-up view of ...

The current commercially available contrast agents are based on polyaminocarboxylate ligands, (Figure 2) which incorporate both nitrogen and oxygen donors.12,26 These ligands are octadentate and leave one coordination site open for binding water molecules. The high ligand denticity is needed to maintain complex stability; but the low number of inner-sphere water molecules compromises relaxivity. Based on the known oxophilicity of Gd3+, ligands containing only oxygen donors with lower denticity should offer high stability with low coordination numbers. This coordination preference of Gd3+ was our guiding principle in designing the tris-bidentate HOPO-based complexes. These complexes incorporate six oxygen donors for Gd3+ binding, leaving two to three coordination sites open for binding inner sphere water molecules. The first HOPO based ligand synthesized comprised of three bidentate HOPO moieties attached via a tris (2-aminoethyl) amine (TREN) cap. The Gd-TREN-1-Me-3,2-HOPO (1) complex has a relaxivity value of 10.5 mM−1s−1 (at 20 MHz), almost twice that of commercial contrast agents. This is most likely a result of the higher number of inner sphere water molecules.27 The X-ray structure of this complex confirms that the hexadentate ligand binds gadolinium through the HOPO oxygen atoms, with two water molecules completing the coordination sphere (Figure 3, left).

Figure 2
Two representative Gd3+ polyaminocarboxylate-based commercial contrast agents with one inner sphere water molecule in each.
Figure 3
X-ray crystal structure of Gd-TREN-1-Me-3,2-HOPO (1) (Left) and the coordination polyhedron (Right). The polyhedron is close to a trigonal bi-capped prism with the missing ligand position pointing directly towards the reader.27

The stability of a Gd-complex is directly related to the concentration of free Gd3+ in vivo and different Gd complexes can be compared in terms of their pM values.28 The Gd-TREN-1-Me-3,2-HOPO complex exhibits a pM value of 20.3 at pH 7.4 (commercial agent Gd(DTPA), Figure 2, has a pM value of 19.09). For a detailed comparison of stability constants of Gd-HOPO-based complexes with current commercial contrast agents the reader is referred to a recent review by our group.29 The low solubility (~ 1 mM) of the Gd-TREN-1-Me-3,2-HOPO complex prompted the synthesis of a series of heteropodal ligands. Ligand cap variation was also explored. The most recent variation was replacing the TREN cap with a triazacyclononane (TACN)-based cap.30 The HOPO-complexes relevant to this review are listed in Figure 4. The Gd-TACN-1-Me-3,2-HOPO complex (4) has solubility as high as 100mM, maintains a reasonably high pM value of 18.7, and has a high relaxivity value of 13.1 mM−1s−1 (20 MHz, 298 K).30

Figure 4
Chemical representations of the Gd3+ hydroxypyridinone complexes discussed in this paper.

Effect of geometry and coordination number on the mechanism of water exchange

The early lanthanide aqua complexes (La3+ −Nd3+) are nine coordinate while the smaller mid to late lanthanides (Gd3+−Lu3+) are eight coordinate; Sm3+ and Eu3+ aqua ions exist in equilibria between eight and nine coordinate species.20 The water exchange process proceeds through either an associative mechanism, or a dissociative mechanism or an interchange mechanism between associative and dissociative (Figure 5). Variable pressure 17O-NMR measurements can be used to measure the volume of activation ([big up triangle, open]V in the equation (δ ln (kex)/ δP)T = −[big up triangle, open]V/RT; kex = water exchange rate) for the exchange process. A large negative [big up triangle, open]V value indicates an associative mechanism; a large positive [big up triangle, open]V value indicates a dissociative mechanism.20 The Gd3+ aqua ion has a small negative volume of activation, indicating an associative interchange water exchange mechanism through a nine-coordinate transition state.31 The commercial polyaminocarboxylate-based ligands have nine coordinate geometries for all lanthanides3,5 and these complexes exchange water through a dissociative process. 31

Figure 5
Representation of the transitions states for the water-exchange processes possible for Gd3+ on the reaction coordinate.20 Associative, interchange, and dissociative mechanisms are depicted in order of increasing volumes of activation.

The X-ray structure of Gd-TREN-1-Me-3,2-HOPO (1) shows an eight coordinate complex which indicates the possibility of an associative water-exchange mechanism.27 The energy difference between the different coordination geometries is very small for lanthanides, and shape measure, S is used to determine which ideal geometry the complex resembles.32 Shape measure is defined as S=min[((1m)Σi=1m(δiθi)2)12], where m is the number of edges, δi is the dihedral angle along the ith edge of the experimental structure (the angle between normals of adjacent faces), and θi is the dihedral angle along the ith edge of the ideal comparison polyhedron. The structure of an eight coordinate lanthanide complex can resemble one of the following high-symmetry polyhedra: the bicapped trigonal prism (C2v), trigonal dodecahedron (D2d) or square antiprism (D4d) (Figure 6). Since S is a true metric, the lowest value of S for the three pairs represents the best fit to the nearest idealized geometry. The crystal structure of Gd-TREN-1-Me-3,2-HOPO (1) is best described as a trigonal dodecahedron. There is an obvious hole in the coordination sphere of the structure, suggesting an open site for binding an additional water-molecule. The crystal structure of Gd-TREN-1,2-HOPO (2) also conforms to an eight coordinate gadolinium with the shape measure suggesting a trigonal bicapped prism (C2v) (Figure 6).33 Since there is a small energy difference between the eight- and nine-coordinate geometries for lanthanides these crystal structures suggest an associative mechanism for water exchange. Additionally, the crystal structure of TREN-1-Me-3,2-HOPO with La3+ has both an eight-coordinate (square antiprismatic) and a nine-coordinate (monocapped square antiprismatic) lanthanide metal center in the unit cell (Figure 7).34 The presence of both eight and nine coordinate lanthanum TREN-1-Me-3,2-HOPO validates that the eight and nine coordinate geometries are indeed very close in energy which should allow facile interconversion between the two coordination states. Finally, the small negative volume of activation ([big up triangle, open]V = −5 ± 1 cm3 mol−1) for Gd-TREN-bis(6-Me-HOPO)-(TAM-TRI) (10) confirms an associative interchange mechanism similar to the gadolinium aqua ion (kex = 10.6 × 108 s−1).35 Analysis of the SBM equations shows that water exchange rates of the order of 108 s−1 are near optimal for attaining high relaxivity values (for field strengths between 20-100 MHz) upon slowing the tumbling motion.5 The small energy gap between the eight and nine coordinate species for the Gd-HOPO complexes, implied by the structures, means that tuning the ligand structure to stabilize the nine coordinate species (q =3) is possible. Since q is directly proportional to relaxivity this should increase relaxivity while maintaining stability. This gives the Gd-HOPO-based complexes an advantage over the commercial polyaminocarboxylate-based complexes, which can only equilibrate between q = 1 (nine coordinate) and q = 0 (eight coordinate) states. 4

Figure 6
A: Crystal structures for Gd-TREN-1,2-HOPO (2) (left) and Gd-TREN-1-Me-3,2-HOPO (1) (right) showing the subtle differences in the coordination polyhedron. B: The ideal polyhedron to which the crystal structures can be compared to by using shape analysis. ...
Figure 7
X-ray Crystal structure of La-TREN-1-Me-3,2-HOPO depicting the eight and nine coordinate La3+ binding sites.34

Measurement of water-exchange rates and implications for development of high relaxivity CA

The water-exchange rates for the Gd-HOPO complexes can be determined by variable temperature proton-decoupled 17O-NMR measurement of the water-nuclear transverse relaxation rates (Figure 8A).34,35 The more soluble heteropodal complexes (6-10) were used since high concentrations are required for these measurements. The data indicate fast water-exchange rates on the order of 108 s−1 (water residence lifetimes τM = 10-15ns); 1 order of magnitude slower than the Gd3+ aqua ion, but 1-2 orders of magnitude faster than the polyaminocarboxylate-based complexes.

Figure 8
Representative plots of the typical relaxometric measurements done for the Gd-HOPO complexes. A: 17O NMR data for Gd-TREN-bis(6-Me-HOPO)-(TAM-TRI) (circles) and simulated curves for GdDO3A (Prohance), a commercial contrast agent (dashed lines).34, 50 ...

An estimate of the water-exchange rate can also be obtained by analyzing the temperature dependence of the longitudinal relaxivity.36 Three different exchange regimes are possible: rapid water-exchange indicated by relaxivity value increase with decreasing temperature (water-exchange rate is not the limiting factor determining relaxivity); slow exchange indicated by relaxivity value decrease with decreasing temperature (water-exchange rate is limiting); and intermediate exchange indicated by little or no dependence of relaxivity on temperature. The Gd-HOPO complexes (6-8, 10) studied showed an exponential increase in relaxivity with decreasing temperature, indicating fast water-exchange with lifetimes (τM) well below 100 ns. 34

Several modifications were made to HOPO-based complexes to increase their water-solubility. Two major routes were: synthesis of heteropodal terephthalamide (TAM) containing ligands bearing water soluble substituents, for example polyethyleneglycol (PEG) chains; and replacing the TREN backbone with more soluble caps. 17O-NMR measurements on these more soluble systems gave a range for water residence lifetimes between 2-50 ns. The Gd-TREN-bis-HOPO-TAM complexes with PEG substituents (11, 12) showed decreasing water exchange rates with increasing PEG chain length (τm = 19 ns for 11 and 31 ns for 12).37 The field dependence of relaxivity, also known as the nuclear magnetic resonance dispersion (NMRD) profile, allows estimation of τM, q and τR when fitted to a model describing the magnetic coupling of the solvent with the paramagnetic agent. Representative NMRD profiles for Gd-HOPO complexes are shown in Figure 8B.38, 39 q values of one were determined by 17O-NMR and NMRD profile for the PEG substituted Gd-TREN-bis-HOPO-TAM complexes (11, 12). The low q values are indicative of partial displacement of the inner-sphere water molecules by the PEG oxygen atoms. Since the water-exchange rate of associatively exchanging complexes depends on the water-concentration, lowering of the local water-concentration caused by the PEG chain leads to a decrease in the water-exchange rate.

Replacement of the TREN backbone with a 2-hydroxymethyl-TREN cap (Figure 4, compound 3) improves the water solubility and keeps the relaxivity similar to that of the original Gd-TREN-1-Me-3,2-HOPO complex.40, 41 Water exchange remains fast and the water residence lifetime is 14 ns. Recently the TREN cap was replaced with a TACN cap which significantly improves the water-solubility, and gives a faster water-exchange rate (τM = 2 ns).30 Molecular mechanics models developed to calculate q from the exposed metal surface area and the strain energy of adding additional inner sphere water molecules predicts a q of 3 for the Gd-TACN-1,2-HOPO complex (5).30, 42 The NMRD profiles and luminescence lifetime measurements on the Eu3+ version of the complex confirms the predictions made by the modeling studies. These results indicate that the TACN-HOPO-based ligands stabilize 9 coordinate gadolinium ions and exchange the inner sphere water molecules by a fast dissociative exchange mechanism. The fast water-exchange rate and the dissociative mechanism of exchange make the TACN-based complexes ideal for attachment to macromolecules to attain high relaxivity values by slowing tumbling rates. A dissociative water exchange mechanism should not depend on local water concentration and hence attachment to macromolecules should not decrease the water-exchange rates as in the case of the PEG substituted TREN complexes.

MR contrast enhancement upon slowing tumbling rates

The optimum water-exchange rate for high relaxivity is field strength dependent. At the current clinically relevant MRI field strength (60 MHz) the optimum value for the water-residence lifetime is 10-30 ns (water exchange rate around 108 s−1). With the advent of new high field (100 MHz and greater) clinical scanners, the optimal water residence time would be around 1ns (water exchange rate close to that of the Gd3+ aqua ion, 109 s−1).29

Commercial Gd3+ polyaminocarboxylate-based complexes have water-exchange rates on the order of 106 s−1, two orders of magnitude slower than the optimum water exchange rate.31 The contrast enhancement that can be obtained upon slowing the motion of these contrast agents is limited. Numerous studies have reported the attachment of the polyaminocarboxylate-based ligands to macromolecules such as PEG chains, polymers, dendrimers and nanoparticles, to slow the tumbling rate of the CA. Moderate enhancements in relaxivities have been observed for these systems, and the interested reader is referred to current reviews.4, 11, 43 A notable result of these studies is that the water-exchange rate usually changes after attachment to the macromolecule. The factors responsible for this are the effect of the macromolecule on the local water concentration near the Gd center and the structural modifications on the ligand required to create the attachment point for the macromolecule. For example, the water residence lifetime of Gd-DTPA (Figure 2) is 300 ns. Upon attachment to polymers to form alkyl-DTPA-bisamide copolymers of the formula [Gd(DTPA-BA)-(CH2)n]x, the water residence times become longer (τM = 2300, 1500, and 2000 ns for n = 6, 10 and 12 respectively).44 For this set of polymers the proton relaxivities were limited by the slow rates of water-exchange. Similarly for Gd-DOTA (Figure 2) (τM = 244 ns), attachment to a lysine based dendrimer increased the water residence time to 1000 ns, limiting the relaxivity to 16.5 mM−1s−1 (20 MHz).

The Gd-HOPO-based contrast agents have water-residence times (τM = 2-30 ns) ideal for current and future high-field instruments. Additionally the water-exchange is not significantly slowed by conjugation to macromolecules. In the case of Gd-TREN-bis-1-Me-3,2-HOPO-TAM based contrast agents with attached PEG chains (compounds 11 and 12), the water residence times increased to 30 ns from 10 ns in the small molecule contrast agent.37, 45 The water-exchange rate was still rapid enough that the relaxivity enhancements obtained from these complexes were not limited by the water-exchange rate. This was also demonstrated by the temperature dependence of relaxivity which decreased with increasing temperature. Attachment of Gd-TREN-bis-1-Me-3,2-HOPO-TAM to a small aspartate-based dendrimer (Compound 13) (molecular weight 1576 g/mol) gave a relaxivity value of 14.3 mM−1s−1 (20 MHz), with 17O-NMR studies indicating optimum water residence time (~10 ns).38 Studies comparing Gd-HOPO complexes to commercial agents, performed by attaching multiple Gd-HOPO-based agents to dendrimers and other macromolecules, have been limited due to the low water-solubility the TREN-based contrast agents. One way to address this solubility issue is to attach the contrast agents to the interior of a soluble macromolecule. If the macromolecule has modifiable interior and exterior surfaces then the exterior surface can be modified with targeting ligands and the interior surface can contain the Gd-HOPO-based agents.

Virus capsids have been recently explored as macromolecular scaffolds for attachment of MRI contrast agents.25, 46, 47 We chose MS2-virus capsids as the soluble protein-based macromolecule to attach our Gd-HOPO-based contrast agents.39, 48 The MS2 capsid shell consists of 180 copies of a coat protein (molecular weight 13.7 kDa) assembled in an icosahedral arrangement.49 The RNA can be removed by treating the capsids with base to yield 27 nm diameter protein shells.48 These capsid shells have 1.8 nm pores through which small molecules and water can access the interior. The size of the capsids is ideal for imaging non-nervous system tissue such as lesions in blood vessels and tumor tissues.39 90 of the modified ligands 14 were attached to the externally modified capsids (Figure 9).48 Attachment of more ligands caused the capsids to precipitate. The same number of ligands were attached to the internally modified capsids for comparison and both the internal and external conjugates were complexed with Gd3+. Relaxivity measurements for the capsid conjugates gave relaxivity values as high as 41.6 mM−1s−1 (per gadolinium; 3952 mM−1s−1 per particle, 30 MHz, 298 K) for the internal conjugates and 30.7 mM−1s−1 (per gadolinium; 2548 mM−1s−1 per particle, 30 MHz, 298 K) for the external conjugates. These high relaxivity values indicate facile water diffusion into the interior of the capsids. The temperature dependence of relaxivity (increase in relaxivity with decreasing temperature) indicates fast water-exchange.

Figure 9
Attachment of Gd-TREN-bis(1-Me-3,2-HOPO)-TAM to bacteriophage MS2 capsids. The crystal structure of the coat protein dimer49 (left) is shown with the modified residues highlighted. Tyr85 (green) in the interior; Lys106 (red), Lys113 (yellow) on the exterior; ...

The NMRD profiles for these conjugates were fitted using the Lipari-Szabo model describing the rotational dynamics.39 This model distinguishes between the two types of motions that influence proton relaxation: the local motion of the contrast agent with a local correlation time τRl and the global motion of the macromolecule with a global correlation time τRg. Additionally there is a parameter (S2) describing the efficiency of the coupling of the local and global motions (S2 = 1 when the CA rotates with the macromolecule and S2 = 0 when the local motion of the CA is totally independent of the motion of the macromolecule). The best fits for the NMRD profiles were obtained with a τM value of 10 ns (values between 2-20 ns were used for the fitting the data) which is optimal for attaining high relaxivities. Although relaxivity values as high as 40 mM−1s−1 were obtained, these values were still lower than the theoretically predicted maximum relaxivity values. These results can be explained if we examine the parameters describing the motion of the system. For both the internal and external conjugates the S2 values were low (S2 = 0.08 for external conjugates and S2 = 0.13 for the internal conjugates at 298 K); with a slightly higher value for the internal conjugates. The τRl value was also higher for the internal conjugates (τRl = 0.44 ns) than for the external conjugates (τRl = 0.31 ns). These results indicate that the rotation of the internally attached contrast agents is more closely linked to the capsid rotation than externally attached agents. This difference seems to translate into higher relaxivity for the internal conjugates. The internal conjugates were obtained by tyrosine modification (Y85 is located on a β-sheet) while the external conjugates were obtained by lysine modifications (K106 and K113 are located on α-helices) and hence internal conjugate flexibility is less than the external conjugate. Theoretical calculations on these systems indicate that higher τRl values (3-15ns), with optimum water-exchange rates can give relaxivity values as high 140 mM−1s−1 at 30 MHz.39 At higher field strengths Gd3+ complexes with faster water-exchange rates will be necessary to attain similar relaxivity values. TACN-based complexes are well suited for this purpose. One of the current goals of the group is to attach the more soluble TACN-based ligands to macromolecules.

Summary and Outlook

Hydroxypyridonate based CAs designed to take advantage of the coordination preferences of Gd3+ have a large number of inner-sphere water molecules and fast water-exchange rates, requisite characteristics for high relaxivity. These ligands have only oxygen donor atoms, resulting in high thermodynamic stability for gadolinium binding. Slowing the tumbling rates of these CAs by attaching them to macromolecules has resulted in high-relaxivity complexes. In moving towards the next generation of high relaxivity contrast agents for molecular imaging these CAs could play an important role provided water-solubility issues are resolved and high kinetic stability for gadolinium binding is also achieved. The ability to attach these contrast agents to the interior of virus capsid shells and the resultant high relaxivity values have opened an avenue for targeted imaging via the attachment of ligands and receptors to the exterior of the capsids. These developments in MRI-CAs in conjunction with improved imaging instruments should help bring targeted imaging closer to the clinic.


We thank all our coworkers and collaborators for their contributions to this project. We thank Prof. Mauro Botta for providing figures of NMRD profiles for this paper. This work was supported by NIH Grant HL69832, by the Director, Office of Science, Office of Basic Energy Sciences, and the Division of Chemical Sciences, Geosciences, and Biosciences of the U.S. Department of Energy at LBNL under Contract No. DE-AC02-05CH11231


1. Lauffer RB. Paramagnetic metal complexes as water proton relaxation agents for NMR imaging - theory and design. Chem. Rev. 1987;87(5):901–927.
2. Merbach AE, Tóth E, editors. The Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging. Wiley; Chichester: 2001.
3. Toth E, Helm L, Merbach AE. Relaxivity of MRI contrast agents. Top. Curr. Chem. 2002;221:61–101.
4. Caravan P. Strategies for increasing the sensitivity of gadolinium based MRI contrast agents. Chem. Soc. Rev. 2006;35(6):512–523. [PubMed]
5. Caravan P, Ellison J, McMurry T, Lauffer R. Gadolinium(III) chelates as MRI contrast agents: Structure, dynamics, and applications. Chem. Rev. 1999;99(9):2293–2352. [PubMed]
6. Ranganathan R, Raju N, Fan H, Zhang X, Tweedle M, Desreux J, Jacques V. Polymethylated DOTA ligands. 2. Synthesis of rigidified lanthanide chelates and studies on the effect of alkyl substitution on conformational mobility and relaxivity. Inorg. Chem. 2002;41(25):6856–6866. [PubMed]
7. Bloembergen N. Spin relaxation processes In a 2-proton system. Phys. Rev. 1956;104(6):1542–1547.
8. Bloembergen N, Morgan LO. Proton relaxation times in paramagnetic solutions effects of electron spin relaxation. J. Chem. Phys. 1961;34(3):842–850.
9. Solomon I. Relaxation processes in a system of 2 spins. Phys. Rev. 1955;99(2):559–565.
10. Solomon I, Bloembergen N. Nuclear magnetic interactions in the HF molecule. J. Chem. Phys. 1956;25(2):261–266.
11. Aime S, Crich SG, Gianolio E, Giovenzana GB, Tei L, Terreno E. High sensitivity lanthanide(III) based probes for MR-medical imaging. Coordin. Chem. Rev. 2006;250(1112):1562–1579.
12. Chan KWY, Wong WT. Small molecular gadolinium(III) complexes as MRI contrast agents for diagnostic imaging. Coordin. Chem. Rev. 2007;251:2428–2451.
13. Frullano L, Meade TJ. Multimodal MRI contrast agents. J. Biol. Inorg. Chem. 2007;12(7):939–949. [PubMed]
14. Hermann P, Kotek J, Kubicek V, Lukes I. Gadolinium(III) complexes as MRI contrast agents: ligand design and properties of the complexes. Dalton Trans. 2008;(23):3027–3047. [PubMed]
15. Major JL, Parigi G, Luchinat C, Meade TJ. The synthesis and in vitro testing of a zinc-activated MRI contrast agent. P. Natl. Acad. Sci. USA. 2007;104(35):13881–13886. [PubMed]
16. Moats RA, Fraser SE, Meade TJ. A “smart” magnetic resonance imaging agent that reports on specific enzymatic activity. Angew. Chem. Int. Ed. 1997;36(7):726–728.
17. Que EL, Chang CJ. A smart magnetic resonance contrast agent for selective copper sensing. J. Am. Chem. Soc. 2006;128:15942–15943. [PubMed]
18. Sherry AD, Woods M. Chemical exchange saturation transfer contrast agents for magnetic resonance imaging. Annual Review of Biomedical Engineering. 2008;10:391–411. [PMC free article] [PubMed]
19. Zhang XA, Lovejoy KS, Jasanoff A, Lippard SJ. Water-soluble porphyrins imaging platform for MM zinc sensing. P. Natl. Acad. Sci. USA. 2007;104(26):10780–10785. [PubMed]
20. Helm L, Merbach AE. Water exchange on metal ions: experiments and simulations. Coordin. Chem. Rev. 1999;187:151–181.
21. Pidcock E, Moore GR. Structural characteristics of protein binding sites for calcium and lanthanide ions. J. Biol. Inorg. Chem. 2001;6(56):479–489. [PubMed]
22. Yang ZR, Batra R, Floyd DL, Hung HC, Chang GG, Tong L. Potent and competitive inhibition of malic enzymes by lanthanide ions. Biochem. Biophys. Res. Comm. 2000;274(2):440–444. [PubMed]
23. Burling FT, Weis WI, Flaherty KM, Brunger AT. Direct observation of protein solvation and discrete disorder with experimental crystallographic phases. Science. 1996;271(5245):72–77. [PubMed]
24. Ng KKS, Kolatkar AR, Park-Snyder S, Feinberg H, Clark DA, Drickamer K, Weis WI. Orientation of bound ligands in mannose-binding proteins - Implications for multivalent ligand recognition. J. Biol. Chem. 2002;277(18):16088–16095. [PubMed]
25. Allen M, Bulte JWM, Liepold L, Basu G, Zywicke HA, Frank JA, Young M, Douglas T. Paramagnetic viral nanoparticles as potential high-relaxivity magnetic resonance contrast agents. Magn. Reson. Med. 2005;54(4):807–812. [PubMed]
26. Aime S, Botta M, Terreno E. Gd(III)-based contrast agents for MRI. 2005;57:173–237.
27. Xu J, Franklin SJ, Whisenhunt DW, Raymond KN. Gadolinium Complex Of Tris[(3-Hydroxy-1-Methyl-2-Oxo-1,2-Didehydropyridine-4-Carboxamido)Ethyl ]Amine - A New Class Of Gadolinium Magnetic-Resonance Relaxation Agents. J. Am. Chem. Soc. 1995;117(27):7245–7246.
28. pGd = − log[Gd]free; [Gd]total = 1 μM, [L]total = 10 μM (pH 7.4, 25°C, 0.1 M KCl)
29. Werner EJ, Datta A, Jocher CJ, Raymond KN. Recent developments in the pursuit of high-relaxivity MRI contrast agents. Angew. Chem. Int. Ed. 2008;47(45):8568–8580. [PubMed]
30. Werner EJ, Avedano S, Botta M, Hay BP, Moore EG, Aime S, Raymond KN. Highly soluble tris-hydroxypyridonate Gd(III) complexes with increased hydration number, fast water exchange, slow electronic relaxation, and high relaxivity. J. Am. Chem. Soc. 2007;129(7):1870–1871. [PMC free article] [PubMed]
31. Powell DH, NiDhubhghaill OM, Pubanz D, Helm L, Lebedev YS, Schlaepfer W, Merbach AE. Structural and dynamic parameters obtained from O-17 NMR, EPR, and NMRD studies of monomeric and dimeric Gd3+ complexes of interest in magnetic resonance imaging: An integrated and theoretically self consistent approach. J. Am. Chem. Soc. 1996;118(39):9333–9346.
32. Xu JD, Radkov E, Ziegler M, Raymond KN. Plutonium(IV) sequestration: Structural and thermodynamic evaluation of the extraordinarily stable cerium(IV) hydroxypyridinonate complexes. Inorg. Chem. 2000;39(18):4156–4164. [PubMed]
33. Xu J, Churchill DG, Botta M, Raymond KN. Gadolinium(III) 1,2-hydroxypyridonate-based complexes: Toward MRI contrast agents of high relaxivity. Inorg. Chem. 2004;43(18):5492–5494. [PubMed]
34. Cohen SM, Xu JD, Radkov E, Raymond KN, Botta M, Barge A, Aime S. Syntheses and relaxation properties of mixed gadolinium hydroxypyridinonate MRI contrast agents. Inorg. Chem. 2000;39(25):5747–5756. [PubMed]
35. Thompson MK, Botta M, Nicolle G, Helm L, Aime S, Merbach AE, Raymond KN. A highly stable gadolinium complex with a fast, associative mechanism of water exchange. J. Am. Chem. Soc. 2003;125(47):14274–14275. [PubMed]
36. Aime S, Botta M, Crich SG, Giovenzana G, Pagliarin R, Sisti M, Terreno E. NMR relaxometric studies of Gd(III) complexes with heptadentate macrocyclic ligands. Magn. Res. Chem. 1998;36:S200–S208.
37. Doble DMJ, Botta M, Wang J, Aime S, Barge A, Raymond KN. Optimization of the relaxivity of MRI contrast agents: Effect of poly(ethylene glycol) chains on the water-exchange rates of Gd-III complexes. J. Am. Chem. Soc. 2001;123(43):10758–10759. [PubMed]
38. Pierre VC, Botta M, Raymond KN. Dendrimeric gadolinium chelate with fast water exchange and high relaxivity at high magnetic field strength. J. Am. Chem. Soc. 2005;127(2):504–505. [PubMed]
39. Datta A, Hooker JM, Botta M, Francis MB, Aime S, Raymond KN. High relaxivity gadolinium hydroxypyridonate-viral capsid conjugates: Nanosized MRI contrast agents. J. Am. Chem. Soc. 2008;130(8):2546–2552. [PubMed]
40. Hajela S, Botta M, Giraudo S, Xu JD, Raymond KN, Aime S. A tris-hydroxymethyl-substituted derivative of Gd-TREN-Me-3,2-HOPO: An MRI relaxation agent with improved efficiency. J. Am. Chem. Soc. 2000;122(45):11228–11229.
41. Hajela SP, Johnson AR, Xu JD, Sunderland CJ, Cohen SM, Caulder DL, Raymond KN. Synthesis of homochiral tris(2-alkyl-2-aminoethyl)amine derivatives from chiral alphaamino aldehydes and their application in the synthesis of water soluble chelators. Inorg. Chem. 2001;40(13):3208–3216. [PubMed]
42. Hay BP, Werner EJ, Raymond KN. Estimating the number of bound waters in Gd(III) complexes revisited. Improved methods for the prediction of q-values. Bioconjugate Chem. 2004;15(6):1496–1502. [PubMed]
43. Barrett T, Kobayashi H, Brechbiel M, Choyke PL. Macromolecular MRI contrast agents for imaging tumor angiogenesis. Eur. J. Radiol. 2006;60(3):353–366. [PubMed]
44. Toth E, Helm L, Kellar KE, Merbach AE. Gd(DTPA-bisamide)alkyl copolymers: A hint for the formation of MRI contrast agents with very high relaxivity. Chem. Eur. J. 1999;5(4):1202–1211.
45. Thompson MK, Doble DMJ, Tso LS, Barra S, Botta M, Aime S, Raymond KN. Hetero-tripodal hydroxypyridonate gadolinium complexes: Syntheses, relaxometric properties, water exchange dynamics, and human serum albumin binding. Inorg. Chem. 2004;43(26):8577–8586. [PubMed]
46. Anderson EA, Isaacman S, Peabody DS, Wang EY, Canary JW, Kirshenbaum K. Viral nanoparticles donning a paramagnetic coat: Conjugation of MRI contrast agents to the MS2 capsid. Nano Lett. 2006;6(6):1160–1164. [PubMed]
47. Prasuhn DE, Yeh RM, Obenaus A, Manchester M, Finn MG. Viral MRI contrast agents: coordination of Gd by native virions and attachment of Gd complexes by azide-alkyne cycloaddition. Chem. Comm. 2007;(12):1269–1271. [PubMed]
48. Hooker JM, Datta A, Botta M, Raymond KN, Francis MB. Magnetic resonance contrast agents from viral capsid shells: A comparison of exterior and interior cargo strategies. Nano Lett. 2007;7:2207–2210. [PubMed]
49. Valegard K, Liljas L, Fridborg K, Unge T. The 3-Dimensional Structure Of The Bacterial-Virus Ms2. Nature. 1990;345(6270):36–41. [PubMed]
50. Aime S, Botta M, Crich SG, Giovenzana G, Pagliarin R, Sisti M, Terreno E. NMR relaxometric studies of Gd(III) complexes with heptadentate macrocyclic ligands. Magn. reson. Chem. 1998;36:S200–S208.