Synthesis Of PEG-Coated, Ultrasmall, Manganese-Doped Iron Oxide Nanoparticles With High Relaxivity For T₁/T₂ Dual-Contrast Magnetic Resonance Imaging

Introduction

In recent decades, magnetic resonance imaging (MRI) has played an irreplaceable role in medical diagnosis because of its efficient soft tissue contrast, high spatial resolution, lack of ionizing radiation, and unrestricted signal penetration depth.^1,2 Particularly, molecular magnetic resonance imaging (mMRI) has extended beyond traditional MRI technology. mMRI can obtain qualitative and quantitative information about biological processes in vivo at the cellular and molecular levels, greatly expanding the research and application fields of MRI.^3–5 In recent years, mMRI has been intensively studied in gene imaging,⁶ cell tracing,⁷ drug screening,⁸ and evaluating early diagnoses and therapeutic effects of disease.^9,10 Contrast agent (CA) serving as the signal source of mMRI is the most important ingredient and has reasonably become a hot spot in the field of mMRI.¹¹

Although many types of CA based on magnetic nanoparticles (NPs) with high longitudinal relaxivity (r₁)¹² or transverse relaxivity (r₂)¹³ have been developed, single-contrast mMRI still possesses insufficient sensitivity and specificity due to its innate properties. Transverse relaxation time (T₂) single-contrast CA usually results in MR signal loss and susceptibility artifacts.¹⁴ Additionally, the darker areas in MR T₂-weighted imaging (T₂WI) induced by the T₂ single-contrast CA may be confused with other physiological or pathological low-signal regions, such as bleeding, calcification, and metal deposition, which will significantly reduce the sensitivity and specificity of mMRI.¹⁵ Similarly, some normal tissues (such as fatty tissue) with hyperintensity in MR longitudinal relaxation time (T₁)-weighted imaging (T₁WI) may be misunderstood as bright lesions enhanced by T₁ CA. T₁/T₂ dual-contrast MRI using the same CA provides complementary information about a lesion to self-confirm and provide fault-free MR images, proving that it can significantly improve the detection accuracy.^16,17

Thus far, multifarious CAs have been reported for T₁/T₂ dual-contrast MRI. A common strategy for the synthesis of T₁/T₂ dual-contrast CA is to integrate T₁ and T₂ signal moieties as hybrid nanoparticles.^17–23 However, due to the magnetic coupling effect, the enhancement generated by the T₁ and T₂ moieties will interact with each other, leading to the compromise of both r₁ and r₂. Embedding the T₁ species into the T₂ species can parallel the electronic spins of paramagnetic T₁ CA with the direction of the magnetic field induced by the T₂ CA, and reasonably enhance both the T₁ and T₂ relaxivity of the T₁/T₂ dual-contrast CA.¹⁶

Because of good biocompatibility, high relaxivity, and easy and versatile functionalization, superparamagnetic iron oxide nanoparticles (IONPs) have been extensively studied and have emerged as promising T₂ CA.²⁴ Benefiting from the size-dependent contrast, the ultrasmall IONPs also have demonstrated satisfactory r₁, making them alternative T₁/T₂ dual-contrast CA.^25,26 Nevertheless, the increased T₁ relaxivity of the ultrasmall IONPs is the result of the consumption of the size, resulting in a relatively small r₂. Thus, ultrasmall IONPs may be not optimal as T₁/T₂ dual-contrast CA.

Divalent manganese ions (Mn²⁺) possess five unpaired electrons and are of higher paramagnetism, and Mn²⁺ doping has been proven to be an effective route to improve the r₁ of ultrasmall IONPs.^12,27,28 Additionally, doped Mn²⁺ with a higher magnetic moment (μ_B=5.92) can occupy both the tetrahedral (T_d) and octahedral holes (O_h) in the crystal lattices, resulting in a special mixed spinel structure, a larger saturation magnetization (Ms), and a high r₂ of manganese-doped IONPs (Mn-IONPs).^2,29 Moreover, based on the embedding rationale, the doped Mn²⁺ and ultrasmall IONPs show synergetic enhancement, which will further improve both r₁ and r₂ of Mn-IONPs. Therefore, Mn-IONPs could be better candidates as dual-contrast CA.

Usually, the hydrophobic inorganic nanoparticles need to be modified with either natural macromolecules or synthetic polymers for hydrophilia and further functionalization for biomedical applications. 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-(methoxy [polyethylene glycol]) (DSPE-PEG), with various advantages such as a low critical micellar concentration (CMC), excellent biocompatibility, and long blood circulation time has been diffusely employed for surface modification. More importantly, it was reported that DSPE-PEG can significantly improve r₂ of the IONPs via the immobilization of nearby water molecules by forming hydrogen bonds.^30,31

Herein, Mn-IONPs were prepared through the dynamic simultaneous thermal decomposition (DSTD) method. Strikingly, after coating with DSPE-PEG, the obtained hydrophilic manganese-doped iron oxide nanocomposites (Mn-IONPs@PEG) showed both good r₁ and r₂ in vitro and in vivo, demonstrating that these nanocomposites comprising ultrasmall Mn-IONPs and DSPE-PEG could be ideal candidates for use as T₁/T₂ dual-contrast CA.

Materials And Methods

Materials

Oleyl alcohol and dibenzyl ether were purchased from Aldrich Chemical Co. (Saint Louis, MO, USA). Absolute ethyl alcohol was obtained from Chongqing Chuandong Chemical Co. Ltd. (Chongqing, China). Hexane and tetrahydrofuran (THF) were purchased from Adamas Reagent Co. Ltd. (Shanghai, China). DSPE-PEG2000 was obtained from Shanghai Ponsure Biotechnology Co. Ltd. (Shanghai, China). Dulbecco’s Modified Eagle’s Medium (DMEM) was purchased from Life Technologies Co. (Saint Louis, MO, USA). Fetal bovine serum (FBS) was obtained from Zhejiang Tianhang Biotechnology Co. Ltd. (Zhejiang, China). Penicillin/streptomycin solution and L-glutamine were purchased from Beyotime Biotechnology Co. Ltd. (Shanghai, China). Cell Count Kit-8 (CCK-8) was purchased from Boster Biological Technology Co. Ltd. (Wuhan, China). All of these reagents were used as received.

Synthesis Of Mn-IONPs

The Mn-IONPs were prepared following a previously reported procedure with modifications.¹² The iron-eruciate and manganese-oleate were first synthesized and described in the supporting information (SI) in detail. Next, 2 mmol of iron-eruciate, 1 mmol of manganese-oleate and 6 mmol of oleyl alcohol were dissolved in 10 mL of dibenzyl ether at room temperature. The mixture was then heated to 100 °C and maintained for 30 min under a constant argon flow and magnetic stirring. Subsequently, the mixture was heated to 265 °C at a rate of 5 °C/min and was maintained for 1 h. After cooling to room temperature, the synthesized Mn-IONPs were washed with ethanol and centrifuged and were finally dispersed in 20 mL of hexane for storage.

Surface Modification

By coating with the DSPE-PEG-2000, the Mn-IONPs were characterized with hydrophilia. Following the previous method,³² 50 mg of Mn-IONPs was first dispersed in 50 mL of THF to form 1 mg/mL of Mn-IONPs solution. Next, DSPE-PEG-2000 was mixed with 1 mL of Mn-IONPs solution with different mass ratios (NPs/PEG = 1:1, 1:5, 1:20, or 1:40) and then was added dropwise into 5 mL of deionized water under ultrasonication. After shaking overnight, the supernatants, including Mn-IONPs@PEG (1:1, 1:5, 1:20, or 1:40), were collected via filtration through a 0.22-μm syringe filter and was stored at 4 °C. Additionally, the hydrophilic Mn-IONPs-TMAH were obtained through ligand exchange with tetramethylammonium hydroxide aqueous solution (TMAH) and were described in the SI in detail.

Characterization

Transmission electron microscopy (TEM; Hitachi HT-7700, Japan) was used to characterize the morphology of the nanoparticles. High-resolution transmission electron microscopy (HRTEM; FEI TECNAI G20, USA) was used to observe the lattice spacing of the nanoparticles. Powder X-ray diffraction (XRD; PANalytical X’Pert Powder, Netherlands) was performed to detect the structure of the nanoparticles. Inductive-coupled plasma optical emission spectrometry (ICP-OES; Thermo Scientific Icap6300 Duo, USA) was carried out to determine the elemental composition of the nanoparticles. The hysteresis loops of the nanoparticles were recorded using a superconducting quantum interference device (SQUID; VSM-Versalab, USA) magnetometer.

Cytotoxicity

The human liver carcinoma cell line HepG2, purchased from the Chinese Academy of Sciences Cell Bank, was used for cytotoxicity analysis in this study. The cells were routinely cultured in DMEM containing 10% FBS, 2 mM L-glutamine, and 100 U/mL penicillin/streptomycin solution in a humidified incubator with 5% CO₂ at 37 °C. CCK-8 was used for the analysis of the cytotoxicity of Mn-IONPs@PEG (1:20). Briefly, the cells were seeded in 96-well plates at a density of 1×10⁴ cells/well and were cultured overnight. Next, fresh mediums containing Mn-IONPs@PEG with serial metal (Fe + Mn) concentrations (0, 1, 5, 10, 20, or 50 µg/mL) were added to replace the previous medium, followed by incubation for an additional 24 h. Each concentration of the samples was repeated five times. Next, the medium was replaced with 100 µL of fresh medium including 10 µL of WST-8 solution. Following 1.5 h of coculture, a spectral scanning multimode reader (Thermo Scientific Varioskan Flash, USA) was used to determine the optical density at the wavelength of 450 nm.

MR Imaging In Vitro

In vitro MR imaging was carried out using a clinical 3.0 T MRI scanner (GE, Signa HDx 3.0T, USA) with a head coil. The Mn-IONPs@PEG and Mn-IONPs-TMAH (shown in SI) solutions with [Fe+Mn] concentrations from 0 to 0.5 mM were imaged using the following parameters: fast spin echo sequence (FSE) T₁WI: repetition time (TR) = 240 ms, echo time (TE) = 7.5 ms, field of view (FOV) = 160×160 mm², matrix = 320×192, slice thickness/spacing =2.0 mm/0.1 mm, number of excitations (NEX) = 4; T₁ mapping: TE = 9 ms, TR = 150, 300, 600, 900, 1200 ms, FOV = 160×160 mm², matrix = 256×256, slice thickness/spacing = 2.0 mm/0.1 mm, NEX = 1; FSE T₂WI: TR = 2000 ms, TE = 46.5 ms, FOV = 160×128 mm², matrix = 192×160, slice thickness/spacing = 2.0 mm/0.1 mm, NEX = 4; T₂ mapping: TR = 1500 ms, TE = 7.8, 15.6, 23.5, 31.3, 39.1, 46.9, 54.8, 62.6 ms, FOV = 160×128 mm², matrix = 160×128, slice thickness/spacing = 2.0 mm/0.4 mm, and NEX = 1.

MR Imaging In Vivo

Animal experiments were conducted in accordance with the National Institutes of Health guidelines on the use of animals in research and were approved by the Laboratory Animal Welfare and Ethics Committee of the Army Medical University. Female C57BL/6 mice with a body weight of approximately 20 g at the age of 8 weeks were purchased from Chongqing TengXin Biotechnology Co. and were housed at the Experimental Animal Center of XinQiao Hospital. The mice were anesthetized with isoflurane and were maintained in a stereotaxic frame (R510IP; RWD Life Science). The T₁WI (TE=9.3 ms, TR=580 ms, FOV = 60×60 mm², matrix = 288×192, slice thickness/spacing =2.0 mm/0 mm, and NEX = 4) and T₂WI (TE=69.6 ms, TR=2820 ms, FOV = 60×60 mm², matrix = 288×192, slice thickness/spacing =2.0 mm/0 mm, and NEX = 6.) Images of mice were acquired using a special rat coil (MS40; Suzhou Medcoil Healthcare Co., Ltd) and a 3.0-T clinical GE MRI scanner before and after injection of the Mn-IONPs@PEG (1:20) with 5 mg [Fe+Mn] per kilogram body weight.

Results

Synthesis And Characterization Of Mn-IONPs

Mn-IONPs were prepared via thermal decomposition of iron-eruciate and manganese-oleate complexes in the presence of oleyl alcohol in dibenzyl ether. Iron-eruciate and manganese-oleate complexes were manufactured through a simple method and are shown in Figure S1. The characteristic bands of iron-eruciate appeared at 1582, 1557, and 1433 cm⁻¹, while those of manganese-oleate were at 1606, 1549, and 1423 cm⁻¹, which were consistent with the data of a previous report.¹² The TEM images and corresponding distribution graphs of the obtained nanoparticles are shown in Figure 1A and B, showing that these nanoparticles were of a size 3.0 ± 0.28 nm with a narrow distribution and good uniformity. The crystal lattice fringes shown in the HRTEM images (Figure 1C and D) demonstrated the high crystallinity of the Mn-IONPs, and the measured interplanar distances of 2.56 and 2.97 Å were equivalent to the (311) and (220) lattice planes of Fe₃O₄, respectively. The XRD patterns are shown in Figure 1E. Compared with the spinel Fe₃O₄ powder diffraction data (JCPDS card no. 01–1111), these diffraction peaks shifted slightly toward that of MnFe₂O₄ (JCPDS card no. 10–0319), which can be attributed to the doped manganese. Coupled with the results of ICP-OES showing that the Mn/Fe molar ratio was approximately 0.20, Mn-IONPs were revealed to be successfully prepared. The hysteresis loop recorded by a SQUID magnetometer using a magnetic field up to 3.0 T at the temperature of 300 K showed that the saturation magnetization (Ms) of Mn-IONPs was 24.45 emu g⁻¹ (Figure 1F). Meanwhile, the negligible coercivity and remanence when the external magnetic field was absent indicate their superparamagnetic behavior. Furthermore, after coating with DSPE-PEG-2000 or TMAH, the obtained hydrophilic Mn-IONPs@PEG or Mn-IONPs-TMAH still maintained size uniformity and showed good stability in water.

Figure 1 (A) TEM image, (B) size distribution graph, (C, D) HRTEM images, (E) XRD pattern, and (F) hysteresis loop of synthesized Mn-IONPs.

Relaxivity Of Mn-IONPs

To assess T₁ and T₂ enhancement, the Mn-IONPs@PEG and Mn-IONPs-TMAH solutions were imaged using a 3.0 T clinical MRI scanner with FSE T₁WI, FSE T₂WI, T₁ mapping and T₂ mapping sequences. The two types of Mn-IONPs solutions all showed concentration-dependent signal enhancement (Figure 2A–D). With the concentration increasing, the signal intensity tended to be brighter in T₁WI but gradually grew darker in T₂WI. The linear fitting of the concentration and 1/T₁ or 1/T₂ showed that r₁ values of Mn-IONPs-TMAH, Mn-IONPs@PEG (1:1), Mn-IONPs@PEG (1:5), Mn-IONPs@PEG (1:20), and Mn-IONPs@PEG (1:40) were 3.7, 2.4, 3.5, 7.1, and 7.8 mM⁻¹s⁻¹, respectively (Figure 2E), and the r₂ values were 36.9, 199.5, 149.7, 120.9, and 94.8 mM⁻¹s⁻¹, respectively (Figure 2F).

Figure 2 In vitro MR images of the Mn-IONPs-TMAH and Mn-IONPs@PEG (1:1, 1:5, 1:20, and 1:40) at 3.0 T: (A) T1-weighted, (B) T1 mapping, (C) T2-weighted and (D) T2 mapping. (E, F) Linear fitting of 1/T1 and 1/T2 over different (Fe+Mn) concentrations of the nanocomposites.

In Vitro Cytotoxicity Assay

Excellent biocompatibility was considered to be an important requirement of Mn-IONPs@PEG for in vivo application. The in vitro cell cytotoxicity assay of Mn-IONPs@PEG (1:20) was performed on HepG2 cells using CCK-8. Figure 3 shows the viability of HepG2 cells co-cultured with the Mn-IONPs@PEG of [Fe+Mn] concentration ranging from 1 to 50 μg/mL at 37 °C for 12 hrs and 24 hrs. All the cells survived under the current experimental conditions. All Mn-IONPs@PEG samples showed insignificant toxicity with a cell viability greater than 80% at a concentration of less than 50 μg/mL for 12 hrs and 24 hrs. These results demonstrated that Mn-IONPs@PEG have a good biocompatibility.

Figure 3 Viability of HepG2 cells after incubation with Mn-IONPs@PEG (1:20) at different concentrations (1–50 µg/mL of [Fe+ Mn]) at 37 °C for 12 hrs and 24 hrs.

In Vivo Contrast-Enhanced MRI

Based on the above results, we further explored the potential of the Mn-IONPs as dual-contrast CA in vivo. Due to the high relaxivity, Mn-IONPs@PEG (1:20) were employed for contrast-enhanced MRI of the liver. Figure 4A shows the axial T₁WI and T₂WI MR images acquired preinjection and postinjection (0, 10, 30, 60, and 120 min) of Mn-IONPs@PEG. The significantly brighter T₁WI MR images and darker T₂WI MR images suggest good r₁ and r₂ values of Mn-IONPs@PEG in vivo. The contrast-enhancement ratios of the MR signal intensity ΔSI (ΔSI = (SI_post − SI_pre)/SI_pre × 100%) are shown in Figure 4B and C. The ΔSI peak of 133% in T₁WI appeared at 10 min after injection followed by a gradual decrease, unexpectedly becoming negative at 120 min. By contrast, the T₂WI images showed obvious enhancement at 10 min after injection but increased to a plateau at 30 min and maintained for a long period.

Figure 4 (A) T1-weighted and T2-weighted MR images of transverse planes of the liver using the FSE sequence acquired at 0, 10, 30, 60, or 120 min after intravenous administration of Mn-IONPs@PEG (1:20). (B, C) Enhancement ratio of the MR signal ΔSI of the liver after contrast-enhancement by Mn-IONPs@PEG (1:20).

Discussion

As is well known, T₁ enhancement of superparamagnetic NPs depends on the interaction of the paramagnetic ions on the surface and nearby proton. Doping paramagnetic ions into the NPs has proven to be an effective route to improve r₁.²⁹ Many kinds of metal ions have been reported to be paramagnetic because of their unpaired electrons. Unlike lanthanide ions, the transition metal ions show both excellent biocompatibility and comparative paramagnetism and should be considered better ions for doping. Mn²⁺ with five unpaired electrons shows the highest paramagnetism compared with other transition metal ions.² Additionally, based on the quantum mechanical inner-sphere theory, the T₁ relaxivity for NPs-based T₁ CA can be given by³³

(1)

where P_M is the mole fraction of metal ions, q is the number of water molecules bound per metal ion, τ_M is the residence lifetime of the bound water, and T_1M, which is related to τ_M and the rotational correlation time (τ_R), is the relaxation time of the bound water protons. It was revealed that the τ_M of Mn(H₂O)₆²⁺ is 4.3 × 10⁴ ns and is 2 orders of magnitude smaller than that of Fe(H₂O)₆³⁺ (5.3 × 10⁶ ns) and other transition metal ions.³³ Thus, Mn²⁺ doping will effectively accelerate spin-lattice relaxation and increase r₁. Moreover, embedding paramagnetic ions into the superparamagnetic NPs significantly boosts both r₁ and r₂ of superparamagnetic hybrid NPs by synergetic interaction between the T₁ and T₂ species.¹⁶

In addition to Mn²⁺ doping, decreasing the size of NPs into the ultrasmall region has also been proven to be an effective approach to boost r₁. The smaller the size of superparamagnetic NPs is, the larger the surface-to-volume ratio is, and the more superficial paramagnetic ions are, and the better r₁ the NPs possess.¹² Thus, sufficiently small Mn²⁺ doping nanoparticles must be synthesized to guarantee good r₁. Theoretically, the key to successfully obtain ultrasmall Mn²⁺ doping nanoparticles is to decompose both the precursors at a similar temperature to form a multiple metal-containing monomer before nucleation to obtain smaller nanoparticles instead of a single-metal-containing monomer, resulting in the formation of a nondopant-containing nucleus initially and then another metal serving as the host to grow nanoparticles.¹² Thus, the Mn-IONPs in this study were prepared by DSTD of the iron-eruciate and manganese-oleate complexes, which had a similar decomposition temperature in the presence of oleyl alcohol in dibenzyl ether instead of iron (III) acetylacetonate and manganese (II) acetylacetonate. From the results of TEM, HRTEM, XRD and ICP, Mn-IONPs with an ultrasmall size of 3.0 ± 0.28 nm were successfully synthesized and the Mn²⁺ was fully proven to be doped successfully.

It is very important to have both good r₁ and r₂ to be good T₁/T₂ dual-contrast CA. However, we decreased the size of the NPs to the ultrasmall range to reach an optimal r₁ of the NPs in this study, resulting in a relative thicker spin-canted layer and a much lower saturation magnetization (Ms) and, finally, a much lower r₂ as expected. Unlike r₁, the relaxation of T₂ CA is dominated by the outer-sphere mechanism and can be given by³⁴

(2)

where γ_I, M_n, ρ, a and L are the proton gyromagnetic ratio, molar mass, density, radius of NPs and impermeable thickness of the surface coating layer, respectively. τ_D is the translational diffusion time, which is determined by r and D according to the equation τ_D = r²/D, where r and D are the effective radius and diffusion coefficient, respectively. It is obvious that r₂ is proportional to a³, while Ms² and τ_D are inversely proportional to L³. To increase Ms, we doped Mn²⁺ into ultrasmall supermagnetic magnetite. Because the doped Mn²⁺ can occupy both the tetrahedral (T_d) and octahedral holes (O_h) in the crystal lattices, Mn-IONPs form a special mixed spinel structure, finally resulting in larger Ms and r₂; this finding has been confirmed in this study by showing that the r₂ of Mn-IONPs-TMAH (36.9 mM⁻¹s⁻¹) is larger than IONPs with the same size (r₂ = 29.2 mM⁻¹s⁻¹).¹⁵ Using another approach by coating Mn-IONPs with DSPE-PEG, we further enhance its r₂.

According to Equation (2), r₂ can be improved by decreasing D. The PEG molecules on the surface of the Mn-IONPs can immobilize surrounding water molecules in a large region by forming hydrogen bonds, resulting in a small D and a large τ_D.^30,31 Thus, it is reasonable that all the DSPE-PEG-coated Mn-IONPs showed a much higher r₂ than THAM-coated Mn-IONPs. However, a decreasing trend for r₂ was noted with the increase in the molar ratio of Mn-IONPs: DSPE-PEG. This phenomenon can be easily explained because the hydrogen–bond interactions between PEG and water protons have a critical molar ratio. Too many coating molecules may exclude water protons from the induced magnetic field from the Mn-IONPs, forming an impermeable coating layer on the surface of the Mn-IONPs and leading to a lower r₂. Interestingly, we also found that r₁ can be changed by varying the coating molecules and the molar ratio of these molecules. Mn-IONPs-TMAH showed a higher r₁ than Mn-IONPs-PEG (1:1) and Mn-IONPs-PEG (1:5) but possessed a lower r₁ than Mn-IONPs-PEG (1:20) and Mn-IONPs-PEG (1:40). Unlike r₂, r₁ increased significantly with the increase in the molar ratio of Mn-IONPs: DSPE-PEG. This phenomenon can be attributed to τ_R and q based on Equation (1). As is well known, T₁ relaxation relies on a chemical exchange between the paramagnetic center and water molecules in the first coordination sphere. The more water molecules located in the first coordination sphere, the better r₁ is. Although NPs become hydrophilic by coating with DSPE-PEG, there is still a DSPE-oleic acid layer that is impermeable on the surface of the NPs, resulting in much fewer water molecules in the first coordination sphere than Mn-IONPs-TMAH, in which the impermeable oleic acid layer has been exchanged. Thus, the r₁ of Mn-IONPs-TMAH was higher than that of Mn-IONPs-PEG (1:1) and Mn-IONPs-PEG (1:5). However, why do Mn-IONPs-PEG (1:20) and Mn-IONPs-PEG (1:40) show a higher r₁ than Mn-IONPs-TMAH? The cause may be that r₁ can also be remarkably influenced by τ_R, which is proportional to the molecular weight. With the increase in the molar ratio of Mn-IONPs: DSPE-PEG, the molecular weight of NPs increases notably, leading to a higher r₁. Over the critical ratio, the shielding effect of the impermeable DSPE-oleic acid layer will be counteracted and τ_R will be dominant, making the r₁ of Mn-IONPs-PEG (1:20) and Mn-IONPs-PEG (1:40) higher than that of Mn-IONPs-TMAH.

Excellent biocompatibility is considered a key point for inorganic NPs application in vivo.¹¹ To improve the biocompatibility, we selected DSPE-PEG, a well-established nonimmunogenic and noncytotoxic material, for Mn-IONPs coating. By CCK-8 analysis, the acquired Mn-IONPs@PEG (1:20) showed negligible cytotoxicity and excellent biocompatibility, which were consistent with the results of previous studies.^30,31,35 Thus, we further explored the potential of Mn-IONPs@PEG (1:20) for T₁/T₂ dual-contrast MR imaging in vivo. Obviously, clear signal intensity changes were noted in both T₁WI and T₂WI over time. The signal intensity changes in the liver may be the result of the biological distribution and metabolism of Mn-IONPs@PEG in vivo. After injection, Mn-IONPs@PEG gradually accumulated in the liver within 10 mins, leading to increased ΔSI of both T₁ and T₂. However, the ΔSI of T₁WI decreased over time because of passive aggregation driven by the phagocytosis of Kupffer cells, resulting in stable and individual Mn-IONPs@PEG gradually aggregating into clusters and causing the larger T₂ contrast enhancement compared with the smaller T₁ contrast enhancement. Particularly, the ΔSI even turned out to be negative at 120 min because the dominated T₂ contrast enhancement significantly affected the T₁ contrast. By contrast, the ΔSI of T₂WI kept increasing until it reached a plateau and there was no downward trend within the observation period because of no degradation. The results obtained from the in vivo MRI verified that Mn-IONPs@PEG are highly sensitive T₁ and T₂ dual CA for imaging of the liver.

Conclusion

In summary, we successfully synthesized 3-nm-sized Mn-IONPs via facile thermal decomposition of iron-eruciate and manganese-oleate complexes in the presence of oleyl alcohol in dibenzyl ether. After encapsulation in DSPE-PEG-2000, the Mn-IONPs@PEG displayed good colloidal stability and excellent biocompatibility. More importantly, due to Mn²⁺ doping and DSPE-PEG-2000 coating, the synthesized NPs showed unexpected high r₁ and r₂. Confirmed by in vitro and in vivo MR imaging, the Mn-IONPs@PEG are ideal candidates for use as T₁/T₂ dual-contrast CA.