Novel hyaluronic acid-coated polymeric micelles with ROS scavenging to encapsulate statins for alleviating atherosclerosis

Inammation and oxidative stress are two major factors that are involved in the pathogenesis of atherosclerosis. A smart drug delivery system that responds to the oxidative microenvironment of atherosclerotic plaques was constructed in the present study. Simvastatin (SIM)-loaded biodegradable polymeric micelles were constructed from hyaluronic acid (HA)-coated poly(ethylene glycol)-poly(tyrosine-ethyl oxalyl) (PEG-Ptyr-EO) for the purpose of simultaneously inhibiting macrophages and decreasing the level of reactive oxygen species (ROS) to treat atherosclerosis. HA coating endows the micelle system the ability of targeting CD44-positive inammatory macrophages. Owing to the ROS-responsive nature of PEG-Ptyr-EO, the micelles can not only be degraded by enzymes, but also consumes ROS by itself at the pathologic sites, upon which the accumulation of pro-inammatory macrophages is effectively suppressed and oxidative stress is alleviated. Consequently, the SIM-loaded micelles demonstrated remarkable therapeutic effects. In conclusion, the SIM-loaded micelle system can synchronically alleviate the accumulation of macrophages and oxidative stress, providing a promising and innovative option against atherosclerosis.


Introduction
Atherosclerosis is a chronic, systemic in ammatory disease of the large and medium-sized arteries, characterized by the development of plaques due to the retention of lipids in the artery wall and in ltration of leukocytes, which can lead to life-threating events such as myocardial infarction and stroke [1][2][3]. In ammation and oxidative stress are two vital factors that are involved in the progression of atherosclerosis [4]. Statins are widely orally prescribed serum low-density lipoprotein (LDL) cholesterollowering drugs, raising the expression of LDL receptor in hepatocytes [5]. Clinical data has shown that orally administered statins can reduce the risk of cardiovascular disease (CVD) and attenuate the development of atherosclerotic plaques [6]. Statins exhibit pleiotropic properties via reductase derived metabolites blockade, including antioxidant effects, anti-in ammatory effects and multiple vasculoprotective effects on endothelial cells, vascular smooth muscle cells and monocytes [7][8][9].
However, considering the low systemic bioavailability, increasing the oral statin dose to attain higher plasma concentrations is not feasible due to the dose-dependent onset of adverse effects such as hepatoxicity and myopathy [3,5]. In addition, it was reported that high-dose statins probably induced body weight loss and abnormal serum biomarkers [10]. All these imperfections hinder the further application of statins in the clinical practice.
Increasing researches have demonstrated that the speci c delivery of statins to in ammatory atherosclerotic plaques might optimize the therapeutic effects [11][12][13][14][15]. Nanocarriers can target plaques by direct in ltration via the injured endothelium or the dysfunctional neovessels in the adventitia [16]. In addition, nanocarriers administered via intravenous or intraperitoneal injection can be endocytosed by circulating phagocytes, followed by translocation to atherosclerotic lesions by cellular recruitment and in ltration. Whereas nanocarriers only serve as vehicles for targeted delivery of different therapeutics to atherosclerotic plaques in most studies, recent progress also indicated that nanocarriers with intrinsic antioxidant and anti-in ammatory activities are promising next-generation therapies for the treatment of atherosclerosis and other in ammatory diseases [17]. Andrographolide-loaded micelles that responds to the oxidative microenvironment of atherosclerotic plaques was constructed based on the block copolymer of poly(ethylene glycol) and poly(propylene sulphide) (PEG-PPS) for the purpose of simultaneously decreasing in ammatory response and the level of reactive oxygen species (ROS) to treat atherosclerosis [4]. Andrographolide exhibits great anti-in ammatory activity and the antioxidant activity of PEG-PPS was endowed by the thioether in the hydrophobic block PPS. Another research has reported a novel dual ROS-sensitive and CD44 receptors targeting nanocarriers composed of oligomeric hyaluronic acid-2'-[propane-2, 2-diyllbls (thio)] diacetic acl-hydroxymethylferrocene (HASF) [18]. Curcumin was encapsulated into HASF and provided potent anti-in ammatory and antioxidant effects. In addition, 2'-[propane-2, 2-diyllbls (thio)] diacetic acl (TKL) and hydroxymethylferrocene (Fc) acted as ROS-responsive linkers and also enhanced antioxidant effects. Another ROS-responsive delivery system combining antiin ammatory and antioxidant effects has been constructed based on arylboronic esters-modi ed amphiphilic oxidation-sensitive chitosan oligosaccharide (Oxi-COS) [19]. Except for the above, peroxalate ester is another ROS-responsive bonds and has been used to establish drug delivery systems [20][21][22].
However, as far as the information goes, it has not been reported that peroxalate esters are used to achieve the combination of anti-oxidation and anti-in ammation in the treatment of atherosclerosis.
Herein, SIM-loaded hyaluronic acid (HA)-coated biodegradable polymeric micelles assembled from an amphiphilic diblock copolymer poly(ethylene glycol)-poly(tyrosine-ethyl oxalyl) (PEG-Ptyr-EO) were established for atherosclerosis treatment. The preparation route of SIM-loaded polymeric micelles was depicted in Scheme 1. SIM-loaded polymeric micelles were prepared by self-assembly and HA coating on the surface of micelles was achieved by electrostatic interaction. The hydrophobic block Ptyr-EO can react with ROS and this process takes place only at the pathologic sites where the ROS concentration is far above that in normal tissues, and thereby the drug delivery e ciency and therapeutic e cacy may be dramatically increased. Furthermore, SIM and the polymers serve to reduce in ammation and ROS levels, respectively, both of which are essential in the formation and progression of atherosclerosis. It is anticipated that, SIM and the drug carrier itself may exert a synergistic effect for effective treatment of atherosclerosis.

Synthesis of Mal-PEG-Ptyr
Mal-PEG-Ptyr was synthesized according to the report [23,24]. Brie y, Mal-PEG-NH 2 (1.0 g, 0.2 mmol) and Tyr-NCA (1.0 g, 4.8 mmol) were dissolved in anhydrous 15 mL DMF. The reaction was performed at 35 o C under N 2 for 72 h. The reaction mixture was concentrated under vacuum and poured into excess cold diethyl ether for ve times. The precipitation was collected and dried under vacuum.

Synthesis of Mal-PEG-Ptyr-EO
Mal-PEG-Ptyr (0.5 g) was dissolved in anhydrous DMF. TEA (0.6 mL) was added into DMF solution. Ethyl oxalyl chloride (0.3 g, 2.2 mmol) in DCM was added dropwise in ice bath. The reaction was stirred for 24 h. After that, the reaction mixture was ltrated and washed by saturated NaCl solution for three times. The organic phase was collected and poured into excess cold diethyl ether for three times. The precipitation was collected and dried under vacuum.

Synthesis of amino-PEG-Ptyr-EO
Mal-PEG-Ptyr-EO (0.3 g) was dissolved in 10 mL DMF with cysteamine hydrochloride (0.1 g) added. The reaction was stirred at room temperature for 24 h. After that, the mixture was diluted with DCM, washed with saturated NaCl solution for three times and dried with anhydrous Na 2 SO 4 . After concentration, the mixture was poured into cold diethyl ether. Amino-PEG-Ptyr-EO was obtained by ltration and drying under vacuum. To label amino-PEG-Ptyr-EO with uorescence dye, 10 mg of Cy5 NHS ester was reacted with 50 mg of amino-PEG-Ptyr-EO to generate Cy5-PEG-Ptyr-EO.

ROS degradation
Mal-PEG-Ptyr-EO was dissolved completely in DMSO-d 6 with addition of 50 mM H 2 O 2 in a nuclear magnetic tube. The tube was incubated at 37 o C for 30 min, 60 min and 120 min. The chemical structure was characterized by 1 H NMR.

Preparation of micelles
Before preparing micelles, the molecular weights and their corresponding distribution of polymers were measured by the gel permeation chromatography system (GPC, Waters). THF was used as the eluent at a low ow rate of 1.0 mL/min. The standard curve was constructed by monodispersed polystyrene standard. The critical micelle concentrations (CMC) for the block copolymers were determined using pyrene as the probe [25]. Brie y, a constant concentration of pyrene (5.0 × 10 − 7 mol/L) was incubated with varying concentrations of lyophilized block copolymer (0.05-51.2 µg/mL) in phosphate buffer solution (PBS) at 37 o C for 36 h. All samples were run with a total sample volume of 1.2 mL in a quartz cuvette. The spectrophotometer (Hengping F9600) was used to record the excitation spectra for each polymer concentration at 338 and 333 nm. A plot of the I 338 /I 333 ratio versus log of the block copolymer concentration was drawn to calculate the CMC.
Micelles based on amino-PEG-Ptyr-EO were prepared by dialysis method. Brie y, 10 mg amino-PEG-Ptyr-EO dissolved in 1 mL DMSO completely was placed into a dialysis tube (M w 3500) and dialyzed against pure water for 24 h. Amino-PEG-Ptyr-EO micelles (PEMs) were obtained by lyophilization. HA-coated PEMs (HPEMs) were prepared by electrostatic attraction. Brie y, PEMs suspension was slowly added to a HA solution under vigorous stirring at room temperature. The mass ration of HA to PEMs and the concentration of HA were further investigated to obtain the optimum formulation. The method to prepare Cy5-labeled micelles was as mentioned above, except that amino-PEG-Ptyr-EO was replaced with amino-PEG-Ptyr-EO/Cy5-PEG-Ptyr-EO (95/5, w/w).

Characterization of micelles
The size distribution and Zeta potential of micelles were recorded by dynamic light scattering (DLS, Malvern Zetasizer). The morphologies of micelles were investigated by transmission electron microscopy (TEM, JEM-2100). In brief, a drop of micelle suspension was placed onto the surface of the copper grid.
After stained for 5 min, the suspension on the copper grid was absorbed by lter paper. The TEM photos were taken after air-drying.

Drug encapsulation and release
SIM-loaded PEMs (SPEMs) were prepared by dialysis method. Brie y, 100 mg amino-PEG-Ptyr-EO and 10 mg SIM were dissolved in 2 mL DMSO completely. Then, the mixture was placed into a dialysis tube (M w 3500) and dialyzed against pure water for 48 h to remove DMSO and unloaded SIM. After that, the suspension was ltrated through 0.45 µm syringe lter. Afterwards, SIM-loaded HPEMs (SHPEMs) were obtained by HA coating with 0.5 mg/mL HA solution with the mass ratio of 3/1 (HA to SPEMs). Besides, SIM-loaded Mal-PEG-Ptyr micelles (SPMs) were prepared by the same method and used in drug release experiments.
To determine the drug loading e ciency (DLE) and drug loading content (DLC), a known amount of SPEMs was dissolved in 1 mL of THF. This solution was then analyzed by an ultraviolet (UV) spectrophotometer at 238 nm with PEMs suspension as the control [26]. The drug concentration in the solution was calculated based on the standard curve. DLE and DLC were obtained in accordance with the following formulae: The drug release was conducted based on SPEMs and SPMs. Brie y, 1 mL of SPEMs or SPMs suspension with the same SIM amount were placed into the dialysis bag (M w 3500) suspended in 20 mL PBS (pH 7.4) containing 0.2% SDS with or without 10 mM H 2 O 2 . The sealed vials were shaken at 37 o C for 72 h. At predetermined time intervals, the release medium was withdrawn to measure the amount of SIM by UV method and 20 mL fresh medium was added. The experiment was performed in triplicates.
2.9 Cell culture RAW264.7 cells were employed to evaluate the cellular uptake and cytotoxicity against SIM and SIMloaded micelles. RAW264.7 cells, highly expressed CD44 receptors, were cultured in Dulbecco's modi ed eagle's medium (DMEM) containing 10% FBS, 100 mg/mL of penicillin and 100 mg/mL of streptomycin at 37 o C in an incubator with 5% CO 2 humid atmosphere. To induce in ammatory reaction, RAW264.7 cells were simulated with 100 ng/mL LPS for 24 h.
2.10 Cellular uptake LPS-simulated RAW264.7 cells were reported to highly express CD44 receptors that can speci cally bind HA ligands. Therefore, RAW264.7 cells were seeded into a 6-well plate with the density of 3.0 × 10 5 cells/well. After incubation for 12 h, the culture medium was replaced with fresh medium containing Cy5labeled PEMs and HPEMs with or without HA pretreatment. After incubation for another 2 h, cells were washed with PBS for two times and xed with 4% paraformaldehyde for 15 min. After stained with DAPI for 10 min, uorescence images of cells were obtained using a uorescence microscopy (CKX53, Olympus).

Cytotoxicity
The cytotoxicity of SIM and SIM-loaded micelles against RAW264.7 cells was measured by MTT assay.
When the cells reached 80% con uence, blank micelles, free SIM, SPEMs or SHPEMs at different SIM concentrations (0.25, 0.5, 1.0, 2.0, 4.0, and 8.0 µg/mL) or different polymer concentrations (50, 100, 200, 300, 400 and 500 µg/mL) in DMEM were added and cultured for 24 h. To evaluate the time-dependent cytotoxicity, cells were incubated with free SIM and SHPEMs (SIM: 4.0 µg/mL) for 24 h, 48 h and 72 h. To survey the in uence of HA coating on the cytotoxicity, cells were incubated with free SIM, SPEMs and SHPEMs at the same SIM concentration (8.0 µg/mL) for 4 h. Then, the culture medium was replaced with fresh medium and cells were incubated for another 20 h. 10 µL of 5 mg/mL MTT solution was added to each well and the cells were incubated for another 4 h. Subsequently, the medium was carefully removed and 100 µL of DMSO was added to each well. The absorbance of each well was recorded on a microplate reader at the wavelength of 562 nm. The experiment was performed in triplicates.
where OD sample means the OD value from wells treated with drugs, OD control from wells treated with DMEM medium and OD blank from wells without cells but culture medium.

ROS generation
DCFH-DA was used to evaluate the generation of intracellular ROS by uorescence imaging and ow cytometry. In brief, RAW264.7 and LPS-induced RAW264.7 cells were seeded in a 6-well plate and cultured for 12 h at 37 °C. Next, cells were incubated with DCFH-DA (10 µM) for 30 min. Then the cells were washed with PBS and incubated with different samples respectively (PBS, free SIM, HPEMs and SHPEMs). After incubation for another 12 h, the cells were washed with PBS to remove free samples, following by observation using a uorescence microscopy and analysis by ow cytometry.

Aminals treatment
Apolipoprotein E-de cient (ApoE −/− ) mice (six-week old) were obtained from the Nanjing University in Nanjing, China. All the animal care and experimental protocols were carried out with review and approval from the Laboratory Animal Welfare and Ethics Committee of the Nanjing University. ApoE −/− mice were fed with a high-fat diet for 8 weeks and treated once a week with saline, free SIM, SPEMs and SHPEMs at a dose of 30 mg/kg for SIM.

ORO and H&E staining
After treatment for 4 weeks, the mice were sacri ced, and the aortas and major organs were harvested. Aortas, from the heart to the iliac bifurcation, were xed by perfusion with paraformaldyhyde (4% in PBS).
After removing the periadventitial tissue, aortas were dissected longitudinally and stained with Oil Red O (ORO) to quantify the lesion areas of the plaques by NIH ImageJ software. Besides, sections of the major organs including heart, liver, spleen, lung, and kidney were also analyzed by hematoxylin-eosin (H&E) staining.

Synthesis of polymers
As shown in Scheme 2, Mal-PEG-Ptyr copolymers were obtained via the ring-opening polymerization of tyrosine N-carboxyanhydride (Tyr-NCA) using Mal-PEG-NH 2 as the initiator (Scheme 1). The 1 H NMR spectrum of Mal-PEG-Ptyr displays the characteristic signals of PEG (peak a, δ 3.56) and phenolic hydroxyl groups (peak f, δ 9.09) (Fig. 1A). Other signals of Ptyr were labeled and attributed to benzene (peak d and e, δ 6.96 and 6.60), methylene (peak c, δ 2.84-2.66) and methyne (peak b, δ 4.43), respectively. The integral area of these signal peaks was calculated and listed in Fig. 1A. The M n of Mal-PEG-Ptyr block copolymers was calculated by comparing the integrals of peak a and peak c, or peak e. The M n of Mal-PEG-Ptyr was also determined by GPC as shown in Table 1 and the trace of GPC was demonstrated in Figure S1. GPC measurements suggested that Mal-PEG-Ptyr had a narrow molecular weight distribution.  Fig. 1B. The emerging signals assigned to ethyl oxalyl esters were listed and labeled as methylene (peak g, δ 4.31) and methyl groups (peak h, δ 1.30) that indicated the successful synthesis of Mal-PEG-Ptyr-EO. The integrals of peak g and h were calculated and compared with these of peak a, suggesting that ca. 20 tyrosine units were modi ed by ethyl oxalyl chloride, which was evidenced by the increase of the molecular weight of Mal-PEG-Ptyr-EO in Table 1.
Notably, after modi cation, the positions of signals assigned to Ptyr had changed (labeled as peak b', c', d' and e'). At last, cysteamine was conjugated with Mal-PEG-Ptyr-EO to give amine-PEG-Ptyr-EO. The 1 H NMR spectrum was demonstrated in Fig. 1C. The emerging peak i (δ 3.17) indicated the successful conjugation.

ROS degradation
It has been reported that peroxalate ester can be oxidized rapidly by H 2 O 2 to generate CO 2 [22]. Hence, the polymer with peroxalate ester linkages serve as a new family of biocompatible and biodegradable materials, which is applied to deliver therapeutic drugs to in ammation sites or tumor tissues that overproduce ROS species. Theoretically, Mal-PEG-Ptyr-EO can be oxidized into Mal-PEG-Ptyr, CO 2 and alcohol as shown in Fig. 2A. To prove it, Mal-PEG-Ptyr-EO was incubated with 50 mM H 2 O 2 for different time and the chemical structures of polymers were monitored as shown in Fig. 2B. After incubation for 30 min, the signal peak h (δ 1.30) turned into peak h' (δ 1.21) and the signal of peak g decreased, indicating the generation of alcohol. After incubation for 60 min and 120 min, peak g disappeared completely. Besides, peak g' was probably hidden by the signal of PEG, which makes it cannot be observed.

Micelle preparation and characterization
Before preparing micelles, the CMC of Mal-PEG-Ptyr and Mal-PEG-Ptyr-EO was determined using pyrene as a uorescence probe. Notably, Mal-PEG-Ptyr-EO displayed a CMC of 3.4 µg/mL, lower than that of Mal-PEG-Ptyr (8.9 µg/mL) as shown in Fig. 3A, which was possibly caused by the modi cation of ethyl oxalyl chloride to improve the hydrophobicity of Ptyr moieties. Micelles based on amine-PEG-Ptyr-EO were prepared by dialysis method. The size distribution and morphology were detected by DLS and TEM in Fig. 3B. It showed an average size of 131.5 ± 6.4 nm with PDI 0.106 ± 0.012. Besides, the micelles exhibited Zeta potential of 38.4 ± 9.7 mV, resulting from free amino groups. HA with molecular weight of 40-100 kDa was coated on the surface of micelles by electrostatic interaction with amino groups. As shown in Fig. 3C, the size increased and the Zeta potential decreased at the beginning with the addition of HA. However, the size decreased rapidly and the Zeta potential continued to decrease when the mass ratio of HA to amine-PEG-Ptyr-EO micelles increased to 3/1, which indicated the complete coating of HA.
To explore the in uence of the concentration of HA, micelles were coated with HA with concentration ranging from 0.5 to 2 mg/mL. As shown in Fig. 3D, with the increase of HA concentration, the average size of coated micelles gradually increased, while the Zeta potential was almost unchanged. Figure 3E showed the size distribution and morphology of HA-coated micelles with the mass ratio of HA (0.5 mg/mL) to micelles of 3/1. It demonstrated uniform spherical micelles with size around 142.3 nm.

Drug loading and release
The in vitro drug release pro les of SPEMs and SPMs in pH 7.4 PBS were evaluated by the dialysis method. As depicted in Fig. 3F

Cellular uptake and cytotoxicity
To investigate CD44-mediated cellular uptake, LPS-induced RAW264.7 cells were incubated with different micelle formulations. Micelles were labeled with Cy5. As shown in Fig. 4A, cells exhibited signi cant Cy5 uorescence after treated with different micelles. Notably, cells treated with Cy5-labeled HPEMs exhibited stronger uorescence than Cy5-labeled PEMs. However, after cells were pre-treated with free HA and then incubated with Cy5-labeled HPEMs, the red uorescence decreased signi cantly. These results indicated the CD44-mediated endocytosis of LPS-induced RAW264.7 against HA-coated micelles.
To study the cytotoxicity of SIM-loaded micelles, the cell viability was tested by MTT assay. Firstly, both PEMs and HPEMs exhibited the cell viability over 90% at all tested concentrations, indicating great biocompatibility ( Figure S2). Both free SIM and SHPEMs showed comparative cell viability of 24 h's incubation. Free SIM showed slightly higher cytotoxicity than SHPEMs at all tested SIM concentrations (Fig. 4B). To survey time-dependent cytotoxicity, LPS-induced RAW264.7 cells were incubated free SIM and SHPEMs with the same SIM concentration (4 µg/mL) for 24 h, 48 h and 72 h, respectively. As shown in Fig. 4C, for SIM and SHPEMs, the cytotoxicity increased with the prolongation of culture time. To further investigate the in uence of HA on the cytotoxicity, cells were incubated with free SIM, SPEMs and SHPEMs at 8 µg/mL for 4 h. After another 20 h's incubation with normal medium, the cell viability was tested ant the results were shown in Fig. 4D. Speci cally, the cell viability of free SIM was 64.1%, slightly higher than that of SHPEMs (57.8%) and signi cantly lower than that of SPEMs (86.7%). These results suggested that HA coating on micelles facilitated cellular uptake and subsequent cytotoxicity.
To investigate the antioxidant effect, cells were treated with different formulations and intracellular ROS was detected using DCFH-DA as the probe. RAW264.7 and LPS-induced RAW264.7 treated with PBS were used as negative and positive control, respectively. As shown in Fig. 4E and Figure S3, ROS increased signi cantly after cells were induced by LPS. After LPS-induced RAW264.7 cells were treated with free SIM, ROS decreased slightly. Notably, ROS decreased dramatically after cells were treated with HPEMs and SHPEMs. Similar results can be observed from DCF uorescence images in Fig. 4F. These results implied that HPEMs can effectively consume intracellular ROS.

Mice treatment
Inspired by excellent cell uptake pro les and the ideal cytotoxicity against macrophages, in vivo research was carried out to evaluate the therapeutic e cacy SHPEMs. AopE −/− mice were fed with a high-fat diet for 8 weeks and then randomly divided into 3 groups (n = 4). Free SIM and SHPEMs were administrated through tail veil for another 4 weeks. Saline was administrated as the control group. After the experiment, the mice were sacri ced and the aortas and major organs were collected. The isolated aorta was stained with ORO shown in Fig. 5A and the plaque area ratio was measured as shown in Fig. 5B. The lesion area ratio of the control group with saline treatment intervention was signi cantly larger than that of other treatment groups. Interestingly, the lesion area of the SHPEMs group was signi cantly smaller than that of the free SIM group, indicating enhanced anti-atherosclerotic effect. The enhanced therapeutic e cacy of SHPEMs was mainly resulted from the targeting ability derived from HA coating. Besides, the histological tissues slides of main organs in different groups were subjected to H&E staining and displayed in Fig. 5C. No obviously pathological changes can be observed in heart, liver, spleen, lung and kidney, suggesting great biosafety and biocompatibility.

Conclusion
In conclusion, oxidative stress and recruitment of pro-in ammatory monocytes are both critical for the development of atherosclerotic plaques in the aorta. SIM possesses excellent anti-in ammatory activity, whereas the poor water solubility and low bioavailability limit its application in vivo. Therefore, an amphipathic block copolymer PEG-Ptyr-EO with oxidation sensitivity was synthesized, assembled into SIM-loaded micelles and coated with HA for targeted atherosclerosis treatment. The SIM-loaded HAcoated micelles showed excellent atherosclerotic plaque-targeted property. The in vitro tests proved that SHPEMs effectively suppressed the growth of RAW264.7 cells and signi cantly reduced the oxidative stress in macrophages. On the other hand, by reacting with ROS, the micellar carrier itself further alleviated oxidative stress in LPS-induced RAW264.7. Consequently, the CD44-targeted and ROS-responsive polymeric micelles loaded with SIM possess excellent therapeutic effects and low side effects in atherosclerosis treatment.