A Multifunctional Integrated Metal‐Free MRI Agent for Early Diagnosis of Oxidative Stress in a Mouse Model of Diabetic Cardiomyopathy

Abstract Reactive oxygen species (ROS) are closely associated with the progression of diabetic cardiomyopathy (DCM) and can be regarded as one of its early biomarkers. Magnetic resonance imaging (MRI) is emerging as a powerful tool for the detection of cardiac abnormalities, but the sensitive and direct ROS‐response MRI probe remains to be developed. This restricts the early diagnosis of DCM and prevents timely clinical interventions, resulting in serious and irreversible pathophysiological abnormalities. Herein, a novel ROS‐response contrast‐enhanced MRI nanoprobe (RCMN) is developed by multi‐functionalizing fluorinated carbon nanosheets (FCNs) with multi‐hydroxyl and 2,2,6,6‐tetramethylpiperidin‐1‐oxyl groups. RCMNs capture ROS and then gather in the heart provisionally, which triggers MRI signal changes to realize the in vivo detection of ROS. In contrast to the clinical MRI agents, the cardiac abnormalities of disease mice is detected 8 weeks in advance with the assistance of RCMNs, which greatly advances the diagnostic window of DCM. To the best of the knowledge, this is the first ROS‐response metal‐free T2‐weighted MRI probe for the early diagnosis of DCM mice model. Furthermore, RCMNs can timely scavenge excessively produced ROS to alleviate oxidative stress.


Preparation of fluorinated carbon nanosheets (FCNs)
300 mg graphene oxide was placed in a closed stainless steel (SUS316) chamber (20 L) equipped with a vacuum line. Firstly, we exchanged internal air and moisture with N 2 three times to exclude their influence for the fluorination process. And then, 80 kPa mixed gas (F 2 /N 2 ) was added into the reactor at room temperature (RT), keeping for a certain amount of time to guarantee GO to react with F 2 by sufficient contact. Furthermore, taking account of paramagnetic behavior and hydrophobicity of 4 FCNs, we prepared three kinds of FCNs (FCN-1, FCN-2 and FCN-3) with different fluorination degree by adjusting fluorination process. After reacting for an hour at RT, the residuary F 2 and generated gases like HF were completely absorbed by absorption tower, and FCN-1 was finally obtained. In contrast, FCN-2 and FCN-3 were fabricated by fluorinating GO two times. After absorption of the redundant gas, subsequent 80 kPa F 2 /N 2 mixed gas was introduced into the chamber again, keeping for 4 h or 9 h. The corresponding products were denoted as FCN-2 and FCN-3 respectively. Additionally, the FCNs with a higher fluorination degree (h-FCN) were prepared by increasing the reaction temperature from RT to 80 ℃. After reacting for an hour at 80 ℃, the residuary F 2 and by-products were disposed according to the above-mentioned procedure, and h-FCN was finally obtained. In addition, Graphene was treated with the mixed gas to fabricate three different fluorinated graphene (FG) by adjusting the fluorination process, and the corresponding products were denoted as FG-1, FG-2, FG-3, respectively.

Preparation of RCMNs
80 mg FCNs (FCN-1, FCN-2 or FCN-3) was dispersed in ethanol (50 mL), and then sonicated for 30 min. Subsequently, the mixture was degassed under continuous argon gas flow. 4-Amino-Tempo and tris(hydroxymethyl)aminomethane (predissolved in deionized water) was separately dissolved in ethanol (15 mL) by different proportions (details see the information in table S1), followed by analogous 5 degassing. Next, reaction system is cooled to about -45 ℃, and the AT solution was added dropwise. After stirring for 1 h, another solution of THAM was added to schlenk bottle gradually, and the reaction was stirred at -45 ℃ for 1 h again.
Afterwards, the functionalized FCNs (fFCNs), namely RCMNs, were separated by high-speed centrifugation, and further purified with alcohol washing more than once until there was no obvious characteristic absorption peak of AT and THAM in the UV-vis absorption spectra of the supernatant. In addition, to eliminate materials with a larger size, centrifugation ran at a low speed (1000 rpm) after repeated washing.
Eventually, the purified fFCNs was dried overnight at RT in vacuum. RCMN-1, RCMN-2 and RCMN-3 respectively corresponded to the functionalized products of FCN-1, FCN-2 and FCN-3. For comparison to free radical-responsive RCMN and excluding electron paramagnetic resonance (EPR) signal interference of potential radicals on graphene sheets, a control probe RCMN-C1 without free radical-responsive moiety was prepared by replacing AT with ATP during the reaction process (details see the information in Table S1).

Material characterization
Fourier-transform infrared spectrum (FTIR) was performed utilizing a Nicolet 560 FTIR instrument in the wavenumber range between 500 and 4000 cm −1 . X-ray diffraction (XRD) patterns were performed on an Ultima IV powder diffractometer (Rigaku Corporation) with a Cu Kα radiation within the 2θ = 5-90° range. The surface chemical composition and valence band spectrum of as-obtained products was examined by X-ray photoelectron spectroscopy (XPS) on a ESCALAB Xi+ spectrometer (Thermo Fisher Scientific, US) accompanied by a monochromatic Al Ka rays (1486.6 eV) under the circumstance of 12.5 kV × 16 mA. For the core-level spectra, the pass energy and step size were set to 30 eV and 0.1 eV, respectively.

7
Energy dispersive X-ray spectroscopy (EDS) and mapping was conducted by Field emission scanning electron microscopy (FESEM) on Nova Nano450. UV-vis spectroscopic analysis was performed on a Shimadzu UV3600 spectrophotometer at RT. Thermal gravimetric analysis (TGA) was performed on Netzsch 209F1 with a heating rate of 10 °C min −1 from 35 to 800 °C under nitrogen atmosphere.
To determine the proton relaxation rates (r 2 ) of fFCNs, the transverse relaxation time T 2 (s) was investigated at different concentrations in 10 mM PBS buffer using a 7.0 T MRI instrument at room temperature. In brief, the fFCNs solutions (0.025, 0.05, 0.1, 1.5, 2 mg mL -1 ) in 10 mM PBS buffer (pH=7.4; n = 3) were placed into 1.5 mL tubes. Then, the transverse relaxation time T 2 (s) were measured at room temperature.
The r 2 was calculated according to the equation represents the concentration of fFCNs, 1/T 2 (0) (s −1 ) is the transverse relaxation rate without paramagnetic species and 1/T 2 (s −1 ) is the transverse relaxation rate with CAs.
EPR measurements were carried out on Bruker EPR EMX Plus (Bruker Beijing Science and Technology Ltd, USA) to capture the radical signals, operating at a frequency of approximately 9.8 GHz using a standard microwave power of 1 mW.

Cell uptake assay
To effectively evaluate cellular uptake of nanoparticles, RCMN-1/RCMN-C1 was loaded with generous fluorescein isothiocyanate Ⅰ (FITC). Firstly, 10 mg RCMN-1/RCMN-C1 was suspended in 10 mL PBS. Next, 10 mg FITC was added to the above solution and mixed with a magnetic stirrer in the dark for 8 h.
RCMN-1/RCMN-C1 loaded with FITC (RCMN-1-FITC/RCMN-C1-FITC) was separated by centrifugation, and further purified with PBS and distilled water washing to remove unbound FITC. Eventually, the product was dried overnight at RT in vacuum and preserved in the dark.
An appropriate amount of RCMN-1-FITC/RCMN-C1-FITC was added to DMEM first, and then the mixed solution was added to the petri dish with adherent cells. After 4 hours of co-incubation, staining and fluorescence photography could be carried out. Rhodamine-Phalloidin respectively.

In vitro cytotoxicity test
To assess cytotoxicity, cardiomyocytes cells, H9C2 cells, were used to evaluate the cytotoxic profiles of RCMN-1 nanoprobes. H9C2 cells were incubated with RCMN-1 at various concentrations in the wells of 96-well culture plates in serum-free culture medium. Following repetitive washing with PBS to remove unbound RCMN-1, the labeled cells were further cultured in fresh cell medium for the appropriate periods of time. Finally, cell proliferation was determined by measuring cell viability with a standard CCK-8 assay, and the cell viability was measured by using a microplate reader (Cyration 3, Biotek, Vermont, USA) at 450 nm at 2 h post CCK-8 addition.
Besides, whether the nanoprobes would induce cell apoptosis was also checked by flow cytometry. Briefly, the H9C2 cells were seeded in 12-well plates (10 5 cells/well) to incubate for 24 h, and then the cells were exposed to the PBS solution of RCMN-1 (100 mM based on fluorine element) for 24 h to measure the apoptosis with flow cytometry.

EPR cell experiment
In the experimental part of EPR in cells level, the ROS assay kit was added to the culture dish in which the mouse cardiomyocytes were incubated, regarded as the experimental group, and the other group was added with the same amount of PBS as the control group. After 4 hours of incubation, trypsin digestion was added. EPR tests were performed after cell technology.

MRI imaging
Animals (mice model) were anesthetized using isoflurane (2% induced concentration and 1% maintained concentration) and scanned under a NOVA 7.0T preclinical horizontal MRI system (Time Medical Systems, Ltd.) by using a mouse heart receive coil. The CMRI protocol is shown in Figure S11c. The results of T 2 mapping and T 2 black blood sequences directly represent the corresponding signal alterations caused by the enrichment of materials.
The selection of the region of interests (ROIs) on the MRI images of the mouse heart is shown in Figure S11b. Since DCM is a diffuse lesion, we measured the average value of the whole myocardium. In order to greatly reduce the experimental error, a reference tube made by the Gd-DTPA solution was introduced in T 2 black blood scans, and the signal-to-back ratio (the ratio of myocardial signal to the standard tube signal) was used as the relative signal value for analysis. After cardiac anatomic localization ( Figure S12a), the T2 black blood and T2 mapping sequences before administration were collected as the baseline. Subsequently, mice were

EPR animal experiment
After the mice were euthanized, the hearts were quickly collected, and the quantitative myocardial tissue was weighed, added with an appropriate amount of RCMN-1 solution in PBS, fully homogenized, and sampled with a capillary glass tube. After measuring the liquid level, the samples were put into the scanning 13 chamber, followed by tuning and eventual EPR scanning (Number of counts: 3000, time constants: 20.48 ms, sweep width: 300 G, sweep time: 12.0 s).

Plasma drug concentration and biodistribution
We used a fluorescence spectrophotometer (722S, Shanghai Yitian Scientific Instruments Co., Ltd. China) to detect the plasma drug concentration. Firstly, RCMN-1-FITC solution (2 mg/mL) was injected into balb/c mice via tail vein at a dose of 4 μL/g, and blood was taken from mice at 2/5/15/30 min or 1/4/8/12/24 h after injection, followed by 3000 r and centrifugation for 15 min to obtain serum samples.
After turning on the machine, the excitation wavelength of FITC (460 nm) was input, and the signal values of the serum at different time points were detected after machine zeroing with blank serum, respectively.
Mice were euthanized 30 minutes or 24 hours after tail vein injection, respectively, and the heart, liver, spleen, lungs, kidneys, brain and muscle were taken, followed by fluorescence imaging using IVIS (Lumina Series III, PerkinElmer Ltd., USA). After fluorescence imaging, individual organ and muscle samples are fixed, embedded, and sectioned, DAPI is added to stain cell nuclei, and finally a fluorescence confocal microscope is used to scan the sections and obtain information on biodistribution.

14
The combustion-ion chromatography method (according to GB/T 41067-2021, China) was adopted to quantify the contents of fluorine. Specifically, the organ tissues of the mice were dried and then ground into powder. Subsequently, 50 mg of the powder was placed in a combustion bottle (an oxygen-rich atmosphere) and burned thoroughly. And then the combustion products were completely absorbed by the NaOH solution (0.02M, 10 mL) at the bottom of the bottle. Finally, the fluorine contents of the samples were tested using ion chromatography (IC, ICS600, Thermo Scientific).

In vivo toxicity evaluation
Female BALB/c mice were divided into two groups (n = 6) and intravenously administrated with RCMN-1 at the dose of 4 μL/g, while the pure PBS was intravenously injected into mice as the negative control. Then, the mice were sacrificed at predefined time points post injection, and the blood was collected for whole blood analysis and serum biochemical analysis. Meanwhile, we administered the same dose of RCMN-1-FITC to mice and euthanized them 0.5/24 h after the injection. And the main organs tissues were collected for fluorescence imaging to study bio-distribution. To further determine the effect of periodic injections of RCMN-1 on the mice, after MRI imaging was completed in the treated and control mice, the mice were euthanized and their organ tissues were prepared for H&E and 15 TUNEL staining. The body weight and survival of the mice were also recorded during the experiment.

Histological analysis
After MRI imaging, the mice were euthanized and their organs were harvested. The TUNEL staining and (hematoxylin and eosin) H&E staining were performed to observe vascular density, cell morphology and apoptosis. Briefly, the heart was fixed in 10% buffered formalin, embedded in paraffin and sliced into 5 μm thickness. The slices were deparaffinized in xylene, dehydrated with graded alcohols, and stained with corresponding kits. Finally, the stained slices were observation under an optical microscope (Zeiss DP80).
For immunostaining, the heart samples were incubated overnight at 4 ℃ with rabbit anti-mouse CD31 antibody (1:200) to identify new vessels.
For ROS staining, fresh heart tissue was frozen and sectioned, then DHE dye was added, followed by DAPI to stain the nuclei, then added anti-fluorescence quenching sealer, and finally observed using a fluorescent confocal microscope (Zeiss DP80).
For blood biochemical analysis, blood was collected by eyeball method and centrifuged to obtain serum, then CREAK, 8-ios-PGEF2α or VEGF kits were added.
Subsequently, quantitative analysis was performed by microplate reader after sufficient reaction.

16
For biochemical analysis of cardiac tissue, after fresh hearts were taken, quantitative myocardial tissue was weighed, fully ground, and then added into SOD, CAT or MDA kits, and quantitative analysis was performed after complete reaction.

Statistical analysis
Statistical analyses were performed using SPSS (Version 23.0) and Prism (Version                represent mean values ± SD, n = 10. Statistical differences were determined by unpaired Student's t-test. NS means no significant difference. **p < 0.01, ***p < 31 0.001. Figure S21. EPR signal of RCMN-1 at different time points after mixing with serum.