Ultrahigh relaxivity and safe probes of manganese oxide nanoparticles for in vivo imaging

Mn-based nanoparticles (NPs) have emerged as new class of probes for magnetic resonance imaging due to the impressive contrast ability. However, the reported Mn-based NPs possess low relaxivity and there are no immunotoxicity data regarding Mn-based NPs as contrast agents. Here, we demonstrate the ultrahigh relaxivity of water protons of 8.26 mM−1s−1 from the Mn3O4 NPs synthesized by a simple and green technique, which is twice higher than that of commercial gadolinium (Gd)-based contrast agents (4.11 mM−1s−1) and the highest value reported to date for Mn-based NPs. We for the first time demonstrate these Mn3O4 NPs biocompatibilities both in vitro and in vivo are satisfactory based on systematical studies of the intrinsic toxicity including cell viability of human nasopharyngeal carcinoma cells, normal nasopharyngeal epithelium, apoptosis in cells and in vivo immunotoxicity. These findings pave the way for the practical clinical diagnosis of Mn based NPs as safe probes for in vivo imaging.

M agnetic resonance imaging (MRI) is a routine diagnostic tool in modern clinical medicine. One of significant advantages of MRI is able to obtain three-dimensional tomographic information about anatomical details with high spatial resolution and soft tissue contrast in a non-invasive and real-time manner [1][2][3][4][5] . In order to compensate the innate low sensitivity, the positive or T 1 contrast agents are employed to increase contrast between organs of interest and normal organs by accelerating the longitudinal relaxivity (r 1 ) of water protons, which leading to a brightening of MR image [6][7][8] . The majority of T 1 MRI contrast probes are currently based on gadolinium (Gd 31 ) in the form of paramagnetic chelates [9][10][11] . However, their uses are occasionally associated with nephrogenic system fibrosis (NSF), which suggests a need of finding alternatives 12-14 . Recently, nanoparticles (NPs) have been extensively used in biomedical application [15][16][17][18][19] . As MRI contrast agents, NPs with high relaxivity and low toxicity are most expected. Among all the candidates, Mn-based NPs are regarded as promising alternatives due to their lower intrinsic toxicity than that of Gd 31 and increasing attention in neuroscience research [20][21][22] . However, the development of Mn-based NPs is hindered by two bottlenecks. One is that the Mn-based NPs with high relaxivity have not been still achieved, e.g., the relaxivity of the reported Mnbased NPs is usually lower than that of the commercial Gd-based agents (4.11 mM 21 s 21 ) [23][24][25] . Another is that there have not been any pre-clinical reports on in vitro and in vivo studies of toxicity of Mn-based NPs [20][21][22][23][24][25] . Nanotoxicity 26 , especially immunotoxicity [27][28][29] , has emerged as one of the critical issues to make NPs into practical clinical applications. Although the standardized assessments on immunotoxicity of NPs in biomedical products have not yet been established, it's essential to assess the immune response to the nanoparticles in the pre-clinical research 30,31 .
Here we synthesize the ligand-free Mn 3 O 4 NPs by a simple and green laser-based technique, i.e., laser ablation in liquid (LAL) [32][33][34][35][36][37][38] . Our measurements indicate that the water proton relaxivity is 8.26 mM 21 s 21 when adding the Mn 3 O 4 NPs, which is twice higher than that of the commercial Gd-DTPA contrast agent (4.11 mM 21 s 21 ) and the highest value reported to date for Mn-based NPs [23][24][25] . We also for the first time take systematically the in vitro and in vivo pre-clinical studies on the toxicity of the as-synthesized Mn 3 O 4 NPs and the pharmacokinetics assays. All the measurements confirm that this Mn-based nanoprobe is safe in biocompatible due to lack of any potential toxicity. Therefore, these results demonstrate that the LAL-derived Mn 3  and safe targeted probes for early tumor diagnosis, and superior to the commercial Gd-based contrast agents in terms of contrast enhancement with a satisfactory biocompatibility.

Results
Structure, morphology and component of Mn 3 O 4 NPs. From Fig. 1a, we can see the uniform and dispersive NPs with a diameter of about 9 nm (calculated from about 200 nanoparticles). The energy dispersive spectrum (Fig. 1b) shows the products are composed of Mn and O elements, and the Cu and C peaks originate from the copper grid and amorphous carbon film support, respectively. The selected area electronic diffraction (Fig. 1c) reveals that these NPs are consistent with strong ring patterns of the tetragonal Mn 3 O 4 structure. The high resolution TEM (Fig. 1d) also confirms this result. The XPS measurements are employed to analyze the Mn oxidation states so as to determine which chemical valence state is responsible for shortening relaxation time. From Fig. 1e, the binding energy of Mn 2p 3/2 peaks components are 641.2 and 642.9 eV, which correspond exactly with the data reported respectively for Mn 21 and Mn 41 39-41 . The splitting of the Mn 3s doublets (Fig. 1f) Fig. 2c and d, the T 1 -weighted MR images clearly show a high contrast enhancement of the xenografted tumor  48 . The same dose of Gd-DTPA is also injected (shown in Supplementary Fig. S5), the signal enhancement is about 23%, which is lower than that of as-synthesized Mn contrast agent (64%). Therefore, both the in vitro and in vivo investigations confirm that the Mn 3 O 4 NPs are more effective than Gd-DTPA in T 1weighted images.
The longitudinal relaxivity is proportional to the hydration number of water (q) that coordinates to the unpaired electrons of contrast agents 9 . Referring to the commercially available clinical contrast agent Gd-DTPA, the ligand DTPA forms a sufficiently stable complex around the Gd 31 ion, and only one coordination site is open up for water ligation, however, the Mn 21 carries five unpaired electrons, which offer more free sites for water ligation and result in higher r 1 9,49 . To further investigate the toxicity of the Mn 3 O 4 NPs in vivo, the immunotoxicity are evaluated in Balb/c mice. In brief, we determine the typical cytokines of innate immune including CD206, CD11b, and CD80/CD86 of monocytes/macrophages in peripheral blood, as well as CD69 cytokine of adaptive immune in lymphocyte cells of peripheral blood and lymph nodes. The results are showed in Fig. 3f and Supplementary Fig. S8. There is significant difference between NPs and positive control groups (LPS), which indicating that the measurement is credible. Though there is statistical difference between Mn 3 O 4 NPs and the negative control groups (PBS) on the expression levels of CD11b, CD206 and LNCD69, which indicates that our nanoprobes do slightly stimulate the immune response system, no obvious difference is found between the Mn 3 O 4 NPs and Gd-DTPA groups. Besides, the blood CD69 of the Mn 3 O 4 NPs group is decreased slightly compared to that of the Gd-DTPA group, which confirming that the as-synthesized Mn-based NPs are as safe as Gd-DTPA. Because Gd-DTPA is the commercial and widely used clinical contrast agent, Mn 3 O 4 NPs might exhibit a little immunotoxicity, but the immune response can be acceptable by body in vivo.
Pharmacokinetics assays including half-time, biodistribution, and excretion. Assessing the toxicity of nanobased biomedicine is involved with physicochemical characteristics. Thus, we first measure the stability of our nanoprobes in blood. The half-life of the Mn 3 O 4 NPs is 63.04 (612.96) min in blood (Fig. 4a), which is much longer than that of Gd-DTPA (20 min) 48 . The longer half-life shows the favorable stability and low blood toxicity in vivo. Importantly, it can effectively improve the accumulation of nanoprobes in tumor tissue during circulation and the sensitivity of MR imaging.
To further investigate the biodistribution and excretion of the Mn 3 O 4 NPs, the quantitative analysis on Mn concentration is measured by inductively coupled plasma mass spectrometry (ICP-MS) in typical organs, xenografted tumor tissues, feces and urine of mice. From Fig. 4b, we can see that our nanoprobes accumulate gradually in the lung, liver, spleen, and tumor tissue, but few are found in the brain, heart, and kidney. The exact concentrations of Mn in different organs are listed in the Supplementary Table S1. Interestingly, the Mn 3 O 4 NPs accumulate increasingly in tumor tissues via the repeated blood circulation, which suggesting that it is a potential tumor-targeting nanoprobe. Moreover, as shown in Fig. 4c, about 50% of Mn is excreted via the hepatobiliary transport system within 1.5 weeks. Though hepatobiliary excretion is a slow process, it can still effectively decrease the occurrence of toxicity due to the accumulation of NPs. Importantly, the biodistribution at the subcellular level is observed by TEM, Fig. 4d shows our nanoprobes are mainly localized in the macrophages in the liver, lung, and spleen, as well as in the cytoplasm of epithelial cells in the xenografted tumor tissue. Since the as-synthesized NPs are dispersed inside the tissues with little aggregation, which leading to gradual excretion and minimal cell toxicity. In addition, no abnormalities are found in histological sections of the main organ including brain, heart, kidney, liver, lung, and spleen ( Supplementary Fig. S9), which suggest that the cellular integrity and tissue morphology are not affected by our nanoprobes.

Discussion
The reason that the r 1 value of the Mn 3 O 4 NP synthesized by LAL is higher than that of other Mn-based NPs is still unclear. We suggest that the distance between water and nanoprobes can be one of the influence factors. The T 1 relaxation of water protons is affected by Mn ion via dipolar mechanism, which is a multifaceted phenomenon. Water in close proximity to ion is relaxed and paramagnetic T 1 relaxation enhancement is a spin-lattice effect, which requires a direct contact between surface Mn ion and water 9,10,39 . Based on the Solomon-Bloembergen-Morgan (SBM) theory 50-53 , a classical existing theory of interpreting relaxation of water protons in the present of contrast agent, the relaxivity has a 1/d 6 dependence on the distance (d) between contrast agents and water proton, which can be simplified as: r 1 / d 26 . So, in this case, the shorter the distance between external Mn ion and water proton is, the higher relaxivity is. Additionally, the surface of the LAL-derived NPs is not blocked by any chemical ligands or residues of any reducing agents, which reduce the distance between Mn ion and water proton. This hypothesis has been verified by changing deionized water into 5 mM SDS solution when ablating the target. The FTIR spectrum exhibits that SDS has coated the surface of Mn 3 O 4 nanocrystals 54,55 , the corresponding relaxivity is dropped to be 1.75 mM 21 s 21 (shown in Supplementary Fig. S3b-3c), which is much lower than the relaxivity of products synthesized in deionzed water (8.26 mM 21 s 21 ). Therefore, clean surface remains when LAL in deionized water, which is likely to result in higher r 1 .
In summary, we have synthesized the Mn 3 O 4 NPs with the ultrahigh relaxivity of 8.26 mM 21 s 21 by a simple and green laser-based technique. We further demonstrate that these Mn-based NPs are safe and effective targeted probes for in vivo imaging based on the in vitro and in vivo assessments of biocompatibility, especially the evidence of immunotoxicity. These findings break through the bottleneck in the application of Mn-based NPs for MRI and pave the way for the practical clinical diagnosis of Mn-based NPs as safe probes for in vivo imaging.

Methods
We stated that all the experiments have been approved by the State Key Laboratory of Oncology in South China of China in this study.

Mn 3 O 4 NPs synthesis.
The details of laser ablation in liquids have been reported in our previous works 33,34 . In this case, a manganese target (99.99% purity) is firstly fixed on the bottom, and then the deionized water is poured into the chamber until the target in covered by 8 mm. Then, a second harmonic produced by a Q-switch Nd: YAG laser device with a wavelength of 10 Hz, and laser pulse power of 70 mJ, is focused onto the surface of manganese target. The spot sized is 1 mm in diameter and the whole ablation lasts for 30 min. The experimental setup is Supplementary shown in Fig. S1. As a result, the brown colloid solution is synthesized and collected into a cuvette. After 24 hours, the upper clear liquid is collected for further measurement.
Products characterization. X-ray diffraction (XRD) was performed with a Rigaku D/ Max-IIIA X-ray diffractometer with Cu Ka radiation (l 5 1.54056 Å , 40 kV, 20 mA) at a scanning rate of 1u s 21 , and transmission electron microscopy (TEM) was carried out with a JEOL JEM-2010HR instrument at an accelerating voltage of 200 kV, equipped with an energy-dispersive X-ray spectrometer (EDS). Sample was ultrasounded for a few minutes and then one drop pipette onto a carbon support film on a copper grid. These techniques are used to identify the structure and morphology of as-synthesized samples. XPS (ESCAlab250) is employed to analyze the composition of the surface of samples. Inductively coupled plasma-atomic (ICP) emission spectrometry using a ThermoFisher iCAP6500Duo has been employed to analyze the concentration of Mn, with an incident power of 1150 W, a plasma gas flow of 14 L/min, and an atomization gas flow of 0.6 L/min.   Half-life in the blood. The half-life in the blood is determined by 30 clean Kunming white mice (50% males and 50% females). Blood is obtained by the tail veins at 5, 15, 30, 60, 120, 180, 240, 360, 480, and 720 min, respectively, after tail vein administration of the Mn 3 O 4 NPs (15 mmol/kg).
Biodistribution at the organ and subcellular level. At the organ lever, brain, liver, lung, spleen, heart, kidney, and tumor are collected at 4, 10, and 24 h, respectively, after nanoprobes injection (15 mmol/kg). At the subcellular level, liver, lung, spleen, and tumor are obtained at 4 h after injection. Samples were measured by TEM.