A heteronuclear ZnGd complex as a potential contrast agent for magnetic resonance imaging

A new ligand H3L, with internal compartments for allocating 3 d metal ions and external donors to bind 4 f ions, was synthesized and completely characterized . R action of H3L with zinc(II) and gadolinium(III) salts allows isolating the heteronuclear complex {[ZnGd(HL)(NO3)(OAc)(CH3OH)](NO3)} ·6H2O (1·6H2O). The ability of 1·6H2O to act as a magnetic resonance imaging (MRI) contrast agent was ev luated and this study shows that both the transversal and longitudinal relaxivities are quite high but the T1/T2 ratio of 7.9 indicates that it could have even greater potential as a T2 contrast agent.


Introduction
The application of gadolinium(III) complexes as contrast agents (CAs) in magnetic resonance imaging requires stable complexation of the metal ion under physiological conditions to avoid the release of the toxic free Gd(III). 1 Several pathways leading to the potential dissociation of Gd 3+ complexes in vivo have been identified: 1) acid-catalyzed dissociation; 2) dissociation catalyzed by endogenous metal ions such as Zn 2+ and Cu 2+ and 3) dissociation assisted by endogenous ligands like citrate, phosphate or bicarbonate.Accordingly, the continuous search for new CAs with increasing stability and relaxivities respect to the commercial ones is still a challenge.In this sense, it should be noted that several gadolinium CAs have shown not only to be stables in the presence of Zn II but to improve their relaxation time. 2,3ith these considerations in mind, in this work we prepare the new ligand H 3 L (Scheme 1), with predesigned compartments to allocate 3d and 4f ions, with the aim of isolating a ZnGd heretonuclear complex and evaluate its ability to act as CA for MRI.

Results and discussion
H 3 L was obtained according to Scheme 1

Scheme 1
Analytical and spectroscopic data corroborate the isolation of the desired compound, with high purity, and unequivocally validate the formation of the imine groups and the imidazolidine ring.In this way, the H4 and H17 protons (Fig. 1) appear as singlets in the 1 H NMR spectrum at 3.83 and 8.16 ppm, respectively, and the IR spectrum shows an intense band at 1633 cm -1 , which can be assigned to ν(C=N).The analytical, spectroscopic and X-ray diffraction characterization of 1•6H 2 O agrees with the structure proposed in Scheme 2. Thus, the comparison of the IR spectrum of the complex with that of the free ligand shows that the band assigned to ν(C=N) splits into two new bands at 1636 and 1648 cm -1 , due to the loss of symmetry of the ligand in the complex.In addition, new bands are observed at 1556 cm -1 , assigned to ν(COO -), and at 1300 and 1285 cm -1 , assigned to NO 3 -vibrations, in agreement with the presence of the acetate and nitrate donors in the metal complex. 4Finally, it is possible to observe a broad band centred at 3264 cm -1 , which indicates the presence of water and methanol in the compound.
Several attempts to grow single crystals of 1•6H 2 O were unsuccessful and only allowed to obtain small single crystals.Thus, X-ray diffraction studies were not of enough quality to solve the structure but they let to know the unit cell parameters, which are the following ones: crystal system: monoclinic; a = 10.82(3),b = 16.29(5),c = 26.67 (7) Å; β = 97.37(6)º; V=4659(37) Å 3 .These parameters are very similar to those found for a ZnDy complex with the same ligand 5 and, therefore, the structure should be very akin and should agree with the proposed one in Scheme 2.
The potentiality of 1•6H 2 O as a MRI contrast agent was evaluated.Thus, longitudinal R 1 and transversal R 2 relaxivities were measured.The R 1 value of 4.90 mM -1 s -1 is comparable with those found for other commercial agents, 6 and, therefore, it could be said that 1•6H 2 O should be a good T 1 contrast agent.Nevertheless, the R 2 value is 38.63 mM -1 s -1 , and, therefore, the T 1 /T 2 (R 2 /R 1 ) ratio is 7.9, value that strongly differs from 1.As a consequence, this precludes the use of 1•6H 2 O as a positive MRI contrast. 7However, the R 2 value is also quite large and the T 1 /T 2 ratio is higher than 6, which is a condition for a species being used as a T 2 contrast agent.Accordingly, it seems that this heteronuclear ZnGd complex could be a candidate as a negative MRI contrast, in spite of usual T 2 agents are based, for example, on magnetic iron oxide nanoparticles. 8he potential of the GdZn system to serve as contrast agent was also demonstrated using in vitro MRI on tube phantoms.The MRI experiments were conducted using a reference without complex (at 12 o'clock in phantoms, see Fig. 2) and six different increasing concentrations (clockwise) of the synthesized complex (0.05, 0.2, 0.5, 1, 1.2 and 1.5 mM).The T 1 -weighted phantom shows that a bright image can be gradually obtained by increasing the complex concentration up to 0.5 mM; in contrast, at higher concentrations the image becomes darker.On the other hand, the T 1 -weighted phantom darkens as the concentration increases.Accordingly, this seems to indicate that, as previously anticipated on the basis of the T 1 /T 2 ratio, 1•6H 2 O has a better performance as a T 2 contrast agent.

Conclusions
The new H 3 L ligand, with differentiated compartments for 3d and 4f metal ions, has been satisfactorily isolated and characterized.The ligand yields the heteronuclear ZnGd complex {[ZnGd(HL)(NO 3 )(OAc)(CH 3 OH)](NO 3 )}•6H 2 O.This complex shows longitudinal relaxivities comparable to those of commercial T 1 CAs but the T 1 /T 2 ratio considerable higher than 1 indicates that it is not a good candidate for being used as a positive MRI contrast agent.Nevertheless, the relatively high transversal relaxivity R 2 and the T 1 /T 2 ratio higher than 6 suggest a better performance as magnetic negative imaging contrast agent.These results are validated by in vitro MRI phamtons.

MRI measurements
MRI studies at high magnetic field were conducted on a 9.4 T MR system (Bruker Biospin, Ettlingen, Germany) with 440 mT/m gradients.A quadrature radio-frequency transmit-receive resonator was used for data acquisition.T 2 -weighted images were acquired using a multi slice multi echo sequence of 11.32 ms echo time, 3000 ms repetition time, 16 echoes, 14 slices, 1 average, FOV of 7.5 cm X 7.5 cm and matrix size of 300 X 300.T 1 -weighted images were acquired using a RARE-VTR sequence of 10.81 ms echo time, rare factor 4, repetition times of 3000 ms, 3235 ms, 3491 ms, 3770 ms, 4078 ms, 4422 ms, 4810 ms, 5256 ms, 5779 ms, 6414 ms, 7221 ms, 8327 ms, 10098 ms and 15000 ms.14 slices, 1 average, FOV of 7.5 cm X 7.5 cm and a matrix size of 300 X 300.Post-processing was performed using ImageJ software (Rasband, W. NIH).