Amphiphilic branched polymeric nitroxides for ecient magnetic resonance imaging of multiple-object in vivo

Background: In order to solve the potential toxicity of metal-based Magnetic resonance imaging (MRI) contrast agents (CAs), a concept of non-metallic MRI CAs Currently, etc.) are being extensively studied because their good stability and imaging mechanism are similar to metal based contrast agents (such as Gd 3+ chelate-based clinical CAs). However, nitroxides lower relaxivity and metabolizes rapidly in vivo are still challenge. Previous researches have proven that the construction of macromolecular nitroxides contrast agents (mORCAs) is a promising solution through the macromolecularization of nitroxides (ie, large molecules carry nitroxides). Macromolecular effects not only improve the stability of nitroxides by limiting their exposure to reductive substances in the body, but also improve the overall 1 H water relaxation by increasing the concentration of nitroxides and slowing the molecular rotation speed. Results: Branched pDHPMA-mPEG-Ppa-PROXYL with high molecular weight (MW=160 kDa) and nitroxides content (0.059 mmol/g), it can form a nanoscale (~ 28 nm) self-assembled aggregate in a water environment and hydrophobic PROXYL can be protected by the hydrophilic outer layer to obtain the strong reduction resistance in vivo. Compared with the small molecular 3-Carboxy-PROXYL (3-CP), Branched pDHPMA-mPEG-Ppa-PROXYL displays there prominent performance: 1) its longitudinal relaxivity (0.50 mM -1 s -1 ) is about three times that of the 3-CP (0.17 mM -1 s -1 ); 2) the blood retention time of nitroxides is increased significantly from a few minutes of 3-CP to 6 h; 3) it could provide long-term and significant enhancement in MR imaging of the tumor, liver, kidney and cardiovascular system (heart and only in tumor, liver and kidney imaging, but also in cardiovascular system imaging, which expands the application scope of this kind of CAs.

Therefore, it has become a research hotspot to overcome the toxicity of metal-based MRI CAs.
In this process, a concept of non-metallic MRI CAs has emerged. Currently, paramagnetic nitrogen oxide radicals (nitroxides) (such as PROXYL, TEMPO, etc.) are being extensively studied because their good stability and imaging mechanism are similar to metal-based contrast agents (such as Gd 3+ chelate-based clinical CAs) [15][16][17][18][19]. However, there are some inherent obstacles of nitroxides that need to be overcame to achieve clinical MRI CAs. First of all, the unpaired electrons (only one) of nitroxides are less than Gd(III) ions (containing seven), so the T1 relaxation efficiency (r1) of nitroxides are significantly lower than that of contrast agents based on Gd(III) ions. In addition, due to the reducible sensitivity, nitroxides will be quickly converted into non-relaxing nitrohydroxyl compounds by the reductive substances in the body [20][21][22][23][24]. And the result is losing paramagnetism, which leads to their short half-life and cannot provide sufficient window time for MR imaging. Previous researches [25][26][27][28][29] have proven that the construction of macromolecular nitroxides-based contrast agents (mORCAs) is a promising solution via the macromolecularization of nitroxides (ie, large molecules carry nitroxides).
Macromolecular effect not only improve the stability of nitroxides by limiting their exposure to reductive substances in the body, but also improve the overall 1 H water relaxation by increasing the concentration of nitroxides and slowing the molecular rotation speed [27,[30][31][32][33]. However, these mORCAs have not obtained ideal results.
It mainly faces two problems: (1) the relaxivities and in vivo imaging time need to be further improved; (2) the biosecurity of macromolecular materials needs to be solved.
For it, in the early stage, we used linear and cross-linked biodegradable PEGylated polyester to construct two novel mORCAs which have made unprecedented achievements in solving the above two problems, and the cross-linked one was the most outstanding [34]. Although the above studies have not completely solved the problems faced by mORCAs, it has injected great expectation and hope into the research field. Therefore, it is necessary to develop more macromolecular carrier materials for constructing more mORCAs in order to find the ideal metal-free MRI CAs.
Among macromolecular carriers, biodegradable poly[N- (1, 3-dihydroxypropyl) methacrylamide] (DHPMA copolymers) are favored due to their great structure controllability, diversified functions, long-term blood circulation, excellent water solubility, non-immunogenicity and good biosafety [35][36][37][38][39]. These copolymers have been successfully employed for the development of metal-based macromolecular CAs (mCAs) [30,40,41]. Therefore, it is expected that the combination of small molecular nitroxide radicals with biodegradable DHPMA copolymers could result in safe and efficient metal-free MRI mCAs. These copolymers help nitroxides to improve the relaxivity in vitro and accumulation of nitroxides in some tissues and organs in vivo, thereby generating a multiplication effect to achieve improvement in the imaging contrast. Meanwhile, these DHPMA copolymers have been demonstrated to display low side effects [42].
In this study, we designed and synthesized a novel nitroxide radicals-based mCA, Branched pDHPMA-mPEG-Ppa-PROXYL, which was derived from an enzyme/GSH sensitive PEGylated branched DHPMA copolymer as the macromolecular skeleton.
Amphiphilic Branched pDHPMA-mPEG-Ppa-PROXYL could self-assembly into nano-sized aggregates in an aqueous environment. As shown in Fig. 1, the hydrophilic components formed the outer layer to encapsulate hydrophobic nitroxides (PROXYL) inside, thereby enhancing the stability of PROXYL. Enhanced stability combined with the macromolecular effects, including long blood circulation, increased nitroxides concentration and slow molecular rotation, rendered this nitroxides-based metal-free mCA to be used for MR imaging of multiple-target in vivo including tumor, liver, kidney and cardiovascular system (heart and aortaventralis). In particular, to our knowledge, this is the first report on nitroxides-based MRI CAs for imaging the cardiovascular system.

Materials and Methods
Materials, methods, synthesis of the PROXYL-based branched biodegradable mORCA (Branched pDHPMA-mPEG-Ppa-PROXYL, Scheme S1) and its characterizations (Table S1) were put in the Supporting Information. In vitro or vivo toxicity, blood compatibility test and cell uptake experiment methods were put in the Supporting Information.

Animal and tumor models
All animals were fed in a control room. 7 × 10 5 4T1 cells were inoculated subcutaneously in the dorsal side of each mouse. When the tumor reached 100 mm 3 , all experimental mice randomly divided for MR imaging and other experimental studies.

In vitro relaxivity and in vivo MR imaging
The  [43] was used to analyze the T1 values of each organ and aortaventralis before and after enhancement.

Results and discussion Preparation and characterization of Branched pDHPMA-mPEG-Ppa-PROXYL
According to our previous studies, Gd(III)-based mCAs constructed from DHPMA copolymers with branched structure generally could make full use of the macromolecular effect to achieve high relaxivities and good in vivo MR imaging.
Therefore, in this study, we prepared a branched DHPMA copolymer whose each branched chain contains an enzyme sensitive GFLG peptide for the construction of a novel biodegradable Branched PROXYL-based mORCA (Branched PCE-mPEG-Ppa-PROXYL, as shown in Scheme S1) in the hope of effectively improving the relaxivity and in vivo MR imaging of PROXYL via significant macromolecular effect. In addition, the detail characterization is listed in Table S1.
Firstly, we prepared a PEGylated PROXYL derivative functionalized by dithiopyridyl (PTE) (PTE-mPEG-PROXYL, as shown in Scheme S1) the way we did before [34], and PTE-mPEG-PROXYL has the following functions: 1) the PTE groups can reacted with thiols (thiol-disulfide exchange reaction) to covalently introduce PROXYL derivatives onto macromolecular material, forming the corresponding mORCA; 2) Due to strong hydrophobility of PROXYL, its introduction will weaken the water solubility of the mORCA, so we adopted PEGylation to solve this problem, in addition, PTE-mPEG-PROXYL contains the same amount of mPEG2000 and PROXYL, so after covalently connecting with the macromolecule, it can achieve a good water-solubility and obtain a higher content of PROXYL, so as to ensure the full play of the multiplier effect; 3) Additionally, PEGylation can further improve MW of the mORCA and enhance its biocompatibility and in vivo stability.
On the other hand, in our previous study, a branched DHPMA copolymer with short enzyme-sensitive GFLG peptide in each branch chain (Branched pDHPMA-SH, as shown in Scheme S1) was prepared via RAFT polymerization induced by VA044 of DHPMA, PTEMA, MA-GFLGK-MA and MA-GFLG-NH-CTA. And the Gd-based mCA derived from this copolymer displayed excellent relaxation performance both in vivo and in vitro, and also good biodegradability and low side effects [42]. Inspired by this result, we selected Branched pDHPMA-SH as macromolecular carrier to construct a novel PROXYL-based mORCA (Branched pDHPMA-mPEG-Ppa-PROXYL, as shown in Scheme S1). The synthesis process was two steps: (1) First, the maleimide-functionalized pyropheophorbide-α (Ppa-Maleimide) as the fluorescent probe was covalently introduced via thiol-ene click chemistry, and the input quantity of Ppa-Maleimide did not exceed 1% of the total amount of raw material, which can not only meet the requirements of fluorescence imaging but also not affect the water solubility of the final copolymer; (2) Next, the PEGylated PROXYL derivative was covalently introduced via disulfide-thiol exchange reaction. The structure of the final copolymer was confirmed by 1 HNMR (Fig. S1), and gel permeation chromatography (GPC) measured its MW as 160 kDa (Table S1). EPR analysis showed that the copolymer had paramagnetism and the spin concentration (nitroxides content) was 0.059 mmol/g (Fig. S2). According to our design ideas, Branched pDHPMA-mPEG-Ppa-PROXYL has amphipathy and can self-assemble into an aggregate with a certain nanometer-size in water phase. For verifying this, we analyzed the particle size and morphology of the copolymer by dynamic light scattering (DLS) and transmission electron microscopy (TEM), and the results showed that Branched pDHPMA-mPEG-Ppa-PROXYL could form a self-assembling aggregate with 28 nm of particle size (Fig. S3-4). The amino acid analysis result (Table S2) showed that the molar ratio of Gly/Phe/Leu was ca. 1.4/1.7/1, which indicated that the short peptide GFLG was introduced into the mORCA. Additionally, from DLS (Fig. S5), the zeta potential of the final copolymer was ca. 0 mV, which indicated that the surface of Branched pDHPMA-mPEG-Ppa-PROXYL was electrically neutral, so it can prevent its adsorption by proteins in the blood, resulting in long retention time in the blood.

Relaxivity of Branched pDHPMA-mPEG-Ppa-PROXYL
A clinical Siemens 3.0 T MRI scanner was used to measure the longitudinal relaxivity

In vivo major organ and aortaventralis imaging
A clinical 3T MRI scanner was used to detect MRI signals of Branched pDHPMA-mPEG-Ppa-PROXYL in living mice to evaluate its suitability as an MRI CA. A small molecule, 3-CP, was employed as a control. As shown in Fig. 3a, the signals in the heart were intensified significantly at 5 min after injection, and the intensified trend continued to reach a peak at 15 min. After that, the signals became slowly weakened, while bright signals were still detectable at 30 min after injection of the contrast agent. Additionally, MR signals in the aortaventralis appeared to be strengthened during the initial 5 min after injection and the signal peak was reached at 5 min. The degree of signal enhancement started to decrease after 5 min, and at 30 min the signal intensity (SI) was observed to reduce to the same level as that before injection (Fig. 3b). In the liver (Fig. 4a), the MRI SI began to increase and reached the enhancement peak at 5 min after injection. After 5 min, the SI began to decline continuously, and recovered at the pre-injection level in 30 min. The SI appeared to increase in the kidney (Fig. 4b)  Furthermore, we performed quantitative analysis of MRI images in the heart, liver, kidney and aortaventralis via the T1 value. In the heart (Fig. 3c), the T1 value increased in the initial few minutes after injection of the mCA and reached the enhancement peak at 15 min. An increase by about 168% in the T1 value was achieved in comparison with the level before enhancement (Fig. 3e). After 15 min, the T1 value gradually decreased. The increasing pattern was also seen for the T1 value in the aortaventralis (Fig. 3d), while the change gradient was much steeper and the value peaked at 5 min with an increase by about 120% compared to the level before enhancement (Fig. 3e). In the liver (Fig. 4c), the T1 value increased significantly within 5 min post-injection of the mCA and peaked at 5 min with the same trend as the MRI SI. The T1 value was enhanced by about 137% compared to the level before enhancement (Fig. 4e). The T1 value then gradually decreased and reduced to the pre-injection level at 30 min. In the kidney (Fig. 4d), the T1 value increased to a peak at 10 min with an enhancement of about 136% compared to the level before enhancement (Fig. 4e). Both MRI SI and the T1 value in the organs (heart, liver and kidney) and aortaventralis displayed negligible changes after injection of 3-CP ( Fig.   S6-S8). This can be explained as: (1)    values for the liver and kidney were quantitatively analyzed (p < 0.05).

In vivo tumor imaging
Based on the above good in vitro and in vivo MR imaging, we further studied the imaging efficacy of Branched pDHPMA-mPEG-Ppa-PROXYL at the tumor site. We also used T1 mapping sequences to scan and observe the MRI enhancement of tumor-bearing mice at various time points. The imaging efficacy was quantified by T1 value, and 3-CP was also set up as a control group for the same scan.
After the injection of the contrast agents, the MRI signal at the tumor site began to strengthen within 5 minutes and reached the enhancement peak (Fig. 5a), and then began to decrease continuously, returning to the pre-injection level at 30 minutes.
Quantitative analysis of the above tumor MRI images by T1 value (Fig. 5b) shows that Branched pDHPMA-mPEG-Ppa-PROXYL reached the enhancement peak of the tumor site at ca. 5 minutes. The T1 value increased by about 144% (Fig. 5c), showing a very high enhancement effect, this result was also consistent with the in vitro relaxivity results. After 5 minutes, the T1 value gradually decreased and returned to the pre-injection level at ca. 30 minutes. Subsequently, small molecular 3-CP group was scanned by the same method as described above, and no significant increase in the signal at the tumor site was found (Fig. S9). It is known that tumor tissue contains higher concentrations of reducing substances (such as GSH) than normal tissues, and this will seriously affect the MR imaging efficacy of Branched pDHPMA-mPEG-Ppa-PROXYL in tumor, but the passive target ability to tumor and the protection of PROXYL by the amphiphilic structure still allowed Branched pDHPMA-mPEG-Ppa-PROXYL to provide good MRI enhancement in tumor tissue. The temporal changes of nitroxides concentrations in the blood of mice were shown in Fig. 6. It was observed that the PROXYL concentrations of 3-CP in the blood rapidly decreased and they were below the detection limit at 1 h after the injection, indicating that the small molecule, 3-CP, was rapidly eliminated by reducing

In vitro cyotoxicty
As shown in Fig. 8a, after incubation with Branched pDHPMA-mPEG-Ppa-PROXYL (from 0 to 1 mg/mL) for 24 h, the viability of 4T1 and HUVEC cells showed no significant difference, and the concentration of PROXYL had negligible impact on the cell viability, which indicated that Branched pDHPMA-mPEG-Ppa PROXYL had no obvious toxicity to 4T1 and HUVEC cells.
This could be due to a neutral surface charge of Branched pDHPMA-mPEG-Ppa PROXYL and a biodegradable macromolecular structure since short peptide linker GFLG in the structure could be cleaved by cathepsin B in the body.

Blood compatibility evaluation
We observed the RBC morphology to evaluate hemolysis induced by  The structure and self-assembly of Branched pDHPMA-mPEG-Ppa-PROXYL.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupportingInformation.doc