Cartilage-targeting Poly(Ethyleneglycol) (PEG)- formononetin (FMN) Nanodrug for Treating Osteoarthritis

Wei Xiong (  xiong91007@163.com ) Guangxi Medical University https://orcid.org/0000-0002-3624-4504 Qiumei Lan Guangxi Medical University Xiaonan Liang Guangxi Medical University Jinmin Zhao Guangxi Medical University Hanji Huang Guangxi Medical University Yanting Zhan Guangxi Medical University Zainen Qin Guangxi Medical University Xianfang Jiang Guangxi medical university Li Zheng Guangxi medical university


Introduction
Characterized by synovial in ammation and cartilage destruction, osteoarthritis (OA) is a chronic and irreversible degenerative disease, which occurs commonly elderly people [1]. Although many clinical therapeutics like nonsteroidal anti-in ammatory drugs (NSAIDs), glucocorticoids (GCs) and other drug treatments are one of the effective strategies for OA [2], there are still some shortcomings need to be overcome such as frequent injection, gastrointestinal, cardiovascular risks, and potential overdose [3].
Thus, it is imperative to nd substitutes that have minimal side effects.
Traditional Chinese medicine has a history of thousands of years, which attracts most attention in recent years. Due to their wide range of pharmacophores, high degrees of stereochemistry [4] and modicum adverse effects [5,6], those natural small molecules extracted from Chinese herbs are regarded as promising drug resources. Formononetin (FMN) is a phytoestrogen puri ed from natural herbal plants (e.g., Astraglus membranaceus, Trifolium pretense). It was reported to have series of pharmacological effects including anti-in ammatory [7,8], anti-angiogenic [9], antioxidant [10], cardioprotective [11], neuroprotective [12] and other effects [13], which has been extensively applicated in treatment of various human diseases, such as diabetic retinopathy, Alzheimer's disease and other diseases [14][15][16]. In recent years, it is found that FMN can effectively decrease proteoglycan loss and improve pericellular matrix formation of chondrocytes [7], which may be developed as potential agents for OA treatment. However, bioavailability of FMN is low because it has poor water solubility and can hardly penetrate through the dense matrix of cartilage [17]. Besides, with lacking speci c targeting, FMN may be rapidly cleared in the joint. Sometimes, the retention time after intra-articular (IA) injection is short [18] and repeated articular injection is inevitable. Thus, it is imperative to increase the water solubility and cartilage-targeting effect of FMN to improve its pharmacological effects.
Polymer-drug conjugates (PDCs) by linking hydrophilic polymers with drug to form uniform sized nanoparticles is one of the effective strategies in drug synthesis to enhance drug solubility and e cacy [19]. Polyethylene glycol (PEG) that is approved by FDA with negligible toxicity and immunogenicity [20,21] has been widely applied in PDCs, which increased drug solubility and improved cell growth [22]. Cho et al. [23] reported that PEGylation with doxorubicin contributed to effective drug delivery, prolonged nanoparticles circulation and signi cantly reduced clearance rate (decreased by 73%) compared with doxorubicin without modi cation. After crosslinking with PEG, Triptolide showed obviously prolonged drug release in plasma, leading to enhanced anti-cancer effect [24]. For cartilagetargeting, peptide-mediated (ect. chondrocyte-a nity peptide CAP, anti-in ammatory peptide KAFAK [25,26] and RGD-modi ed delivery systems [27] have been used for modi cation of free drug molecules. Rothen uh et al. found the ligand WYRGRL (a Coll-II α1chain-binding peptide, CollBP) can speci cally bind to collagen IIα1 that is speci c for cartilage matrix [28]. An in vitro study reported that WYRGRLdexamethasone conjugate exhibited increased cartilage-targeting ability, leading to enhanced antiin ammatory effects as compared with dexamethasone alone [29].
Based on FMN that is the natural anti-in ammatory agent, we synthesized a nanosized amphiphilic polymer drug conjugate (PEG-CollBP-FMN, PCFMN) for OA therapy, which was prepared by PEGylation of FMN followed by coupling with cartilage-targeting peptide (CollBP) to increase the bioavailability of FMN ( Figure. 1). Then we investigated the anti-in ammatory effects of PCFMN on IL-1β stimulated chondrocytes and OA rat joints, as compared with unmodi ed FMN. This study may provide a new insight for the design of a novel agent in OA therapy.

Characterization Of Pcfmn
The morphology of PCFMN was determined by using transmission electron microscopy (TEM) (Hitachi, Japan). The size distribution of PCFMN in aqueous suspension was measured by using Malvern Zetasizer Nano ZS90, and all measurements were carried out at 25 °C. The PCFMN was con rmed by ultra-violet and visible spectrophotometer (UV-Vis, TV-1901, Beijing), FTIR spectrum (PerkinElmer, Spectrum100, USA) and 1 H NMR (AVANCE III HD600, Zurich, Switzerland). Speci cally, to study the solubilization and stability of PCFMN, 5 mg FMN and 5 mg PCFMN were dissolved in 1 mL of ultrapure water, respectively, followed by ultrasound for 15 minutes and station for 3 day, and then photographed for observation.

Culture Of Chondrocytes
Primary chondrocytes were harvested from 5-day-old Sprague Dawley rat. Curtly, cartilage was obtained from the joints of rats' limbs and cut into small pieces of nearly 1 mm size. The cartilage debris were digested by 0.25% trypsin at 37 o C for 30 minutes, and 0.2% type II collagenase (Solarbio, Beijing, China) for 4 h. Cells were collected by centrifugation (1000 rpm, 5 min) and cultured in Modi ed Eagle's medium (Solarbio, Beijing, China) containing 10% (v/v) fetal bovine serum (Sijiqing, Zhejiang, China) and 1% (v/v) penicillin/streptomycin (Solarbio, Beijing, China) in an incubator with 5% CO 2 at 37 °C. Cells were passaged when reaching nearly 80%-90% till second-generation for further research.

Cytotoxicity Studies
The cytotoxicity of PCFMN was measured by MTT assay. Brie y, the chondrocytes were seeded in 96-well plate at a density of 6,000 cells per well. After cells adherence, 200 µL of FMN or PCFMN containing formononetin of different concentrations incubated for 24 h. Then, 15 µL of MTT was added to each well. After another 4 hours incubation, all the medium was removed and then DMSO (150 µL) were added to each well. Finally, the optical density at 490 nm was measured by a microplate reader (Thermo Scienti c Multiskan GO Microplate Spectrophotometer). Also, MTT assay was used to detect the effects of FMN and PCFMN on the proliferation of chondrocytes induced by IL-1β. The chondrocytes were cultured in a 24-well plate with 1.5 × 10 4 cells per well for 24 h, then stimulated with IL-1β (10 ng/mL) to constructed osteoarthritis model in vitro. Finally, treated with the FMN (1.25 µg/mL) or PCFMN (1.25 µg/mL) for another 24 h. Next steps were similar with MTT method.
In vitro cellular uptake As the nanoparticles carry red uorescence dye for cell membrane, cellular uptake of PCFMN and FMN can be detected by uorescence inversion microscope (Olympus, Japan). Chondrocytes were cultured in 24-well plates (1 × 10 4 cells/well). When cells were fully attached, fresh medium containing PFMN-DID (without cartilage-targeting peptide) (10 µL, 1 mg/mL) and PCFMN-DID (10 µL, 1 mg/mL) uorescent nanoparticles were added as a substitute for the old medium. After 24 hours incubation, cold PBS and 4% paraformaldehyde were respectively used to wash cells and then x for 20 minutes. Following, chondrocytes nuclear were stained by DAPI for 15 minutes. Finally, the plate was analyzed by a uorescence inversion microscope (Olympus, Japan) to obtain corresponding uorescence images.
In vitro anti-in ammatory activity The second antibody FITC-anti-rabbit IgG (Boster, Wuhan, China) was then added for incubating with samples one more hour at room temperature in dark. Finally, the samples were treated with DAPI for nuclei stained. Images was captured using a uorescence inversion microscope (Olympus, Japan).

Qrt-pcr Detection
Chondrocytes were seeded in 6-well plates to obtain total RNA. After treated with FMN (1.25 µg/mL) or PCFMN (1.25 µg/mL) treatments for 24 h, the mRNA expression level of cartilage-speci c maker (e.g., Col2a1) and some gene expressions of catabolic markers of OA (e.g., IL-1β, MMP-13 and MMP-3) were analyzed by the real-time quantitative polymerase chain reaction (qRT-PCR). An RNA isolation kit (Megentec, Guangzhou, China) was used to harvest the RNA according to its manufacturer's protocol. The reverse transcription was conducted by using a reverse transcription kit (Fermentas Company, USA). All qRT-PCR reactions were performed by a light Cycle 96 system (Roche, Switzerland) under the conditions of 10 min at 95 o C, followed by 40 cycles of 10 s at 95 o C and then 60 s at 60 o C. The 2 −ΔΔCt method was used to calculate the relative gene expression levels. The primers of RT-qPCR are summarized in Table 1. In vivo optical imaging To analyze the nanodrug retention time in vivo, we used an in-vivo Multispectral Imaging System (Bruker, Germany) to detect the uorescence signals of PCFMN and PFMN labeled with DID in knee OA model.

Anti-in ammatory effect in vivo
After the treatment for 4 and 8 weeks, all rats were euthanized with excessive anesthesia and their knee joint samples were obtained. The repaired articular cartilages were harvested. Blinded to the treatment groups, three independent observers (SC, PYF, XF) conducted the macroscopic evaluation and a nine-area grid of each medial and lateral tibia plateau was used to evaluate the grade of articular cartilage surface (scale of 0-8) [30]. Moreover, the repaired knee joints obtained from the rats were collected and xed with 4% formaldehyde. Then, after decalci ed for one month, the joints were embedded in para n and sliced into thickness of 5 µm. With sections dewaxed, histological analysis (H&E, Safanin O and immunohistochemistry staining for MMP-13) was conducted. Then, images were captured on optical microscope. The OARSI cartilage OA histopathology grading system was adopted to grade the repaired tissues [31].

Statistical analysis
All the data were reported as the means ± standard deviations from at least three repeated experiments.
Comparison between OA and treatment groups was examined by one-way analysis of variance (ANOVA). P < 0.05 was considered statistically signi cant.

PCFMN synthesis and characterization
The synthetic approach for the PCFMN based on amide and esteri cation reaction is outlined in Figure. 2a. Homogeneous distributed nanospheres of PCFMN were observed through TEM, and DLS analysis showed its average diameter of 218 nm ( Figure. 2b and c). The successful synthesis of PCFMN was further con rmed by UV-Vis and FTIR, as shown in Figure. 2d. The UV spectra of the PCFMN showed the typical absorbance peaks from both PEG-CollBP and FMN, indicating that these two components successfully interacted with each other. In FTIR spectra of PCFMN ( Figure.

Retention Time Of Nanodrugs In Oa Joints
In-vivo imaging system was used to detect the retention times of PCFMN and PFMN in the OA joints on day 0,1,3,7 and 19. As shown in ( Figure. 5a and 5b), at different time points, the uorescence intensity in PCFMN group was higher than that in the PFMN group. However, PFMN showed drops in the uorescence intensity with time past and no obvious uorescence was observed on day 19.

PCFMN attenuated the progression of OA in vivo
After treated with PCFMN, rats were sacri ced at 4 and 8 weeks respectively. We conducted macroscopic scoring, hematoxylin-eosin (H&E) staining, Safranin O-fast green staining and immunohistochemistry staining of MMP-13.
Compared with the sham group, OA characteristics, including cartilage erosion, osteophyte formation and deterioration over time were observed in saline-treated group ( Figure. 6a). After 4 and 8 weeks administrating with PCFMN, reducing osteophytes and lesion surface was observed and scores declined to 66.67% and 78.26% respectively. Although FMN also ameliorated the cartilage destruction caused by ACLT, they performed poorer than PCFMN.
Following, we used hematoxylin-eosin (H&E) and Safranin O/fast green (SOFG) staining to assess the cartilage tissues of the different groups. Compared with the saline group, the other two groups presented various degrees of improvement such as proteoglycan retention, no bone erosion and tidemark integrity promotion. Speci cally, less severe lesion, decreased surface denudation, as well as increased tissue cellularity and cloning were observed in PCFMN group, which indicated that the PCFMN group had more importance in chondroprotective effects ( Figure. 7a). Furthermore, the Safranin O/fast green positive staining in the PCFMN group was signi cantly stronger than other groups ( Figure. 7a). All these results indicated that PCFMN group resulted in better glycosaminoglycan deposition, cartilage matrix depletion attenuation and overall cartilage thickness retention. OARSI score was illustrated in Figure. 7b, the PCFMN group, one of the treatment groups, grasped lower OARSI scores with about 54.54% and 90.07% reduction at week 4 and 8, respectively, compared with the PBS group. Also, by means of immunohistochemistry staining, the expression of MMP-13 was also investigated. Positive staining (dark brown) on the cartilage surface representing MMP-13 was observed in OA group ( Figure. 7c). In contrast, the MMP-13 expression of PCFMN group showed signi cant decreases than that of FMN group and approximately identical to sham control.

Discussion
Targeting nanodrugs has emerged as promising therapies for OA treatment, but penetration through the dense cartilage matrix and drug retention in the joint cavity still challenged the therapeutic effects [32]. In this study, we fabricated cartilage-targeting nanodrugs of PCFMN by modi cation of FMN by both PEGylation and coupling with cartilage-targeting peptide. Based on the vitro and vivo tests, it was found that PCFMN could speci cally target to cartilage, which have been proved by our previous research [33] and exert strong anti-arthritic and chondro-protective effects.
PCFMN showed high drug solubilization than unmodi ed FMN after PEGylation ( Figure. 2g), demonstrating the effectiveness of PEGylation. PEGylation can enhance the pharmacokinetic properties of drug [34]. With excellent hydrophilic ability, PEG chains grafted on nanoparticle generate a su ciently thick hydrated cloud [35] that strongly prevents NPs from aggregation. Functionalized with low molecular weight PEG, nanomaterials have enhanced penetration ability through mucus [36] and other tissues, which facilitated drug delivery to the lung [37], breast [38] and colorectal carcinomas [39]. CollB-peptide has been demonstrated to bind to collagen II α1 and shows outstanding application in drug delivery and imaging [28,40]. As evidenced by the cellular uptake analysis ( Figure.  PEGylation and targeted peptide modi cation make PCFMN with good anti-in ammatory and chondroprotective effects both in vitro and in vivo ( Figure. 4-7). As a plant natural product, FMN has antioxidant properties and shows neuroprotective effects against traumatic brain injury through activation of Nrf2dependent antioxidant pathways [41]. Further, FMN showed anti-in ammatory capacity by suppressing IL-6 and TNF-α in neuroin ammatory rats [16] and effectively antagonized proteoglycan loss through decreasing the expression MMP-3, MMP-13 and attenuating oxidative stress [7]. Further, with the studying of the repression of OA-marked factors, such as IL-1, MMP-3 and MMP-13, and the promotion of cartilage-speci c marker (collagen II α1), anti-in ammatory and chondro-protective characters of PCFMN were exhibited ( Figure. 4a). Both MMP-13 and MMP-3 can result in degradation of variety cartilage ECM components in OA processes [42] and MMP-13 exhibited preference for collagen II [43]. As expected, compared with free FMN, PCFMN, a cartilage-targeting and PEGylated nanodrug, had more remarkably effect on down-regulation MMP-13 and preventing cartilage degradation ( Figure. 7), indicating that a nano-targeted drug design is a promising therapeutic strategy for OA.
In conclusion, the PCFMN nanoparticles with improved hydrophobic drug solubilization and cellular uptake ability increased drug accumulation and nally achieved effective anti-arthritic effects. Our study indicates that modi cation of drug by PEGylation and coupling with targeting peptide might be a promising strategy for OA therapy. indicates P < 0.05, **, ## indicates P < 0.01, ***, ### indicates P<0.001. * is the statistical difference between the treatment group and the normal group, and # is the statistical difference between the pairwise comparison among the treatment group.).