2.1 Preparation and Characterization of TUG
A one-step solvothermal method is proposed to prepare ultrasmall paramagnetic iron oxide with stable surface modification of sodium citrate (Fig. 1). Due to a large amount of sodium citrate on the surface of nanoparticles, USPIO has a strong charge repulsion force, which is beneficial to maintain excellent colloidal stability in different liquid environments29, 30. The UG NPs were obtained by the cation exchange between Gd(III) and sodium citrate on the surface of USPIO. Subsequently, carboxyl groups covered the nanoparticles and amino groups on the transferrin were bonded via an EDC/NHS reaction31, 32, resulting in the final product TUG.
The photograph of TEM reveals that the prepared ultrasmall iron oxide nanoparticles are uniformly dispersed with a stable diameter of about 2 nm (Fig. 2a). After addition of Gd ions and Tf, the obtained TUG formed by the assembly of ultrasmall iron oxide shows good dispersion with a diameter of about 50 nm (Fig. 2b). SEM element line scanning analysis of TUG clearly shows three elements of iron, gadolinium, and oxygen homogeneously co-present in the obtained nanoprobe (Fig. S1). Figure 2c shows the zeta potentials and hydrodynamic sizes of the USPIO, UG, and TUG. In detail, the zeta potential of USPIO is -40.0 mV due to the abundant carboxyl groups, while after addition of Gd(III) and Tf, the zeta potential increased to -20.0 mV and − 18.0 mV, respectively, indicating successful surface modification (Fig. 2c). The hydrodynamic sizes of USPIO, UG, and TUG are 2 nm, 50 nm, and 82 nm, respectively, which also implying the formation of nanoprobe with successful surface modification of transferrin. Furthermore, the results of zeta potentials and hydrodynamic sizes of TUG in different media (H2O, PBS, and RPMI) for 7 days indicate that TUG can maintain colloidal stability for a long time, which is suitable to in vivo application (Fig. S2). During the synthesis process, the feed ratio of Fe/Gd was set to 10:1, and the actual ratio of Fe/Gd in TUG measured by ICP-OES was 16.7:1. Additionally, ICP-OES results show that the molar ratio of Fe/Na in USPIO nanoparticles and TUG is 3.7 and 61.3, respectively, confirming that sodium ions have been successfully exchanged by Gd ions (Fig. 2d). The conjugation of transferrin was confirmed by FTIR and TGA (Fig. 2e and Fig. 2f). After modification with transferrin, the absorption peaks at 3390 cm− 1, 2930 cm− 1, 1652 cm− 1, and 584 cm− 1 correspond to the O-H, C-H, N-H, and Fe-O stretching of TUG, indicating the successful modification of Tf. The TGA curve (Fig. 2f) shows a modest decline at the beginning of the test implying a 10% weight loss before 100 oC caused by the adsorbed water in TUG. The total weight loss from 200 oC to 500 oC due to the thermal decomposition of organics is around 55%, in which the amount of Tf can be calculated to be 37.5%. Additionally, the stability experiment of TUG verifies that less than 5% free gadolinium ions released from TUG after 8 hours in the saline (Fig. 2g), which ensures its hypotoxicity.
2.2 T1 Relaxometry and MRI Imaging
The r2 and r1 values of USPIO and TUG can be calculated by plotting the reciprocal of relaxation time (1/T2 or 1/T1) as a function of Fe concentration (Fig. 2h and Fig. 2i). It is found that, for USPIO, the r1 and r2 values are 0.1 mM− 1s− 1 and 0.6 mM− 1s− 1, respectively, with the r2/r1 ratio of 6.0. Interesting, for TUG, the r1 and r2 values are 3.2 mM− 1s− 1 and 4.1 mM− 1s− 1, respectively, with the r2/r1 ratio of 1.3. Compared with USPIO, TUG shows higher r1 and lower ratio of r2/r1, which could be ascribed to the shortened longitudinal relaxation time due to the synergistic effect of the obtained cluster structure.
2.3 Phagocytosis, Cytotoxicity and in vitro MRI of TUG in 4T1 cells
The cytocompatibility of the contrast agents is the premise of applying TUG in vivo. As shown in Fig. 3a, after 24 h co-culture, TUG shows negligible cytotoxicity to 4T1 cells (the cell viability of 4T1 cells is above 90% within the given concentration from 0-100 µg/mL). It is worth mentioning that when the co-incubation time prolonged to 48 h under iron content of 100 µg/mL, the cell survival rate remained above 80%, indicating good cytocompatibility of prepared TUG. Furthermore, we validated TUG MRI performance at the cellular level before performing in vivo animal imaging. As shown in Fig. 3b-c, regardless of TUG group or commercial contrast agent Magnevist (Gd-DTPA), the MR signal of 4T1 cells after co-incubation increases with increasing concentration of different contrast agents. Interestingly, at the same concentration, the MR signal of the TUG group is remarkably higher than that of the commercial contrast agent group, indicating that TUG has better imaging performance than the commercial contrast agent. Besides, the introduction of transferrin increases the cell
phagocytosis of nanoparticles due to the overexpressed transferrin receptors of 4T1 breast cancer cells. We thus evaluated the difference in phagocytosis of TUG and UG by ICP-OES and Prussian blue staining. As shown in Fig. 3e, at the same dose, the phagocytosis of TUG by 4T1 cells reaches 36.1 μg per 100,000 cells, while the phagocytosis of non-targeted nanoparticles is only 8.3 μg per 100,000 cells. The amount of nanoparticles consumed by cells in TUG group was 4.3 times of the UG group and the statistical results proved that the modification of transferrin considerably increases the phagocytosis of cells by nanoclusters. The results of Prussian blue staining show that 4T1 cells treated with TUG show clear blue color, while almost no blue signals appeared in the saline-combined UG group, which was consistent with the ICP results (Fig. 3d-e).
2.4 MRI of TUG Nanoprobes in vivo
The enhancement effect of TUG as an MRI contrast agent in vitro was studied by comparing and analyzing T1-weighted MRI images (Fig. S3). The MRI of TUG shows that T1 signal increases with the increase of the Gd proportion. Then, MR images of mice were obtained for further study of the
enhancement effect of MRI (Fig. 4 and Fig. S6). T1-weighted MR images of whole tumor areas were acquired before and 1 h and 8 h after injection by recording the tumor area signal values from all periods and drawing average luminance histogram of MR image tumor area. In the subcutaneous tumor model, the signal peak was detected in mice injected with Magnevist 10min after injection, since Magnevist only passively accumulates in the peripheral vascular tissues of tumor sites by EPR effect, resulting in a low concentration of TUG in the tumor region, and almost unchanged relaxation time of regional hydrogen proton T1. Eight hours after the injection of TUG, the MR images of the tumor areas of the mice are not only brighter than those before the injection, but also brighter than those during the peak in the tumor areas of the mice injected with Magnevist. In the tumor models in situ, Magnevist's signal peak time is about 30 min, and the magnitude of Magnevist's signal peak occurs in less than 0.5 hours after the injection of TUG. It is proved that the T1-weighted imaging effect of TUG is superior to Magnevist in tumor regions. Notably, the peak time varies in different tumor models, which could be ascribed to the fact that the blood supply of subcutaneous tumor is not as rich as that of breast tumors in situ, leading to longer peak time.
2.5 Safety Evaluation of TUG
The long-term biosafety of TUG in living mice after intravenous injection was also evaluated. Section pictures of organs stained by Hematoxylin and eosin (H&E) showed that physiological morphologies of organs including heart, lung, liver, spleen, and kidney of Balb/c female mice remained normal for 14 days after injection of TUG even at a high dose of 15 mg/kg (Fig. 5). To determine whether TUG hurts liver function and renal function, the levels of enzymes and chemicals in serum such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN) and creatinine (CRE) were measured and no abnormals found. Also, the hematocrit, hemoglobin, mean corpuscular volume, mean corpuscular hemoglobin, lymph cells, platelet, red blood cells, red cell distribution, and white blood cells remained almost unchanged after injection of TUG. All the above results indicate high biosafety of the prepared TUG.