Intra-articular injection of kartogenin-conjugated polyurethane nanoparticles attenuates the progression of osteoarthritis

The polymers were synthesized from poly-(ethylene glycol) (PEG), hexamethylene diisocyanate (HDI) and N-BOC-Serinol with molar ratio of 1:2:1, followed by the further deprotection process of BOC-protected amino groups, which was performed as previously described method (Fang et al., 2014). Briefly, under nitrogen protection, 2 g PEG (1000 Da; 2 mmol) and 0.67 g HDI (4 mmol) were mixed in 20 mL DMSO in a three-necked flask, followed by addition of 0.05 wt% Sn(Oct)z. The reaction was carried out at 80 °C for 3 h and then cooled at room temperature. 0.382 g N-BOC-Serinol (2 mmol) in DMSO solution were dropwise added to the prepolymer solution. The final concentration of polymer solution was 5% (w/v) and the reaction continued at 80 °C for 18 h with stirring. After that, the mixture was precipitated in diethyl ether, and the resulting polymers were purified by dissolving in chloroform and further precipitating with diethyl ether for three times, dried in a vacuum oven at 45 °C for 2 days to obtain PB. To generate amphiphilic polyurethane with pendant amino groups (PN), synthesized PB (3 g) were dissolved to a 50% (w/v) concentration in 6 mL anhydrous chloroform/trifluoroacetic acid (TFA) (50/50) in a round bottom flask and stirred at room temperature for 1 h to remove the BOC-protected groups. After reaction, the excess anhydrous chloroform and TFA were moved through rotary evaporation. The polymers were further purified by dissolving in chloroform and precipitating with diethyl ether for three times. After that, the precipitates were dissolved and neutralized in 2% (w/v) NaHCO₃ aqueous solution (pH = 8.3) to make sure TFA remove clearly. Then purified by dialysis (Mw = 3500 Da) against deionized water, and lyophilized (Figure 1(A)).

PN was grafted KGN through N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC)/N-hydroxysuccinimide (NHS) condensation reaction to synthesize PN-KGN. First, NHS and EDC were added to KGN in DMSO to activate carboxylic acid groups (molar ratio of KGN: EDC: NHS is 3:6:2). After 30 min, the PN/DMSO solution was dropwise added to KGN solution, followed by stirring at room temperature for 24 h in dark. Then the products were dialyzed (Mw = 3500 Da) and lyophilized for 24 h (Figure 1(B)).

Both Fourier transform infrared spectroscopy (FTIR) and proton nuclear magnetic resonance spectroscopy (¹H NMR) were used to characterize the surface chemistry of the synthesized PN-KGN. The lyophilized powders of PN-KGN were applied on the FTIR sample folder and recorded on Nicolet 6700 FTIR spectrometer (Thermo Scientific, Waltham, MA). ¹H NMR spectra were obtained using an Avance 400 NMR spectrometer (Bruker, Fällanden, Switzerland). The mean diameter and PDI of nanospheres were determined by dynamic laser light scattering (DLS, B1-200SM, Brookhaven, NY). The morphology of nanospheres was observed by transmission electron microscope (TEM, JEM-2100F, Jeol, Japan) operating at 200 kV.

The PN-KGN nanospheres (10 mg) were placed in 3 mL saline at 37 °C in a shaking incubator (100 rpm). The saline was collected after centrifugation (14,000 rpm, 15 min) and replaced with saline at each sampling time point. The amounts of released KGN in the collected saline were measured by reverse-phase high-performance liquid chromatography (HPLC; 600E-2487, Waters, Milford, MA) spectrum using a C-18 column (150 × 4.6 mm, 5 µm). The analysis was carried out with a flow rate of 1.0 mL/min and recorded at 274 nm for 30 min. The calibration curve for KGN was established in the range 2–200 mg/L.

Primary chondrocytes were obtained from Sprague Dawley (SD) rats. The articular cartilage derived from the terminal of femur and tibia was digested with 0.2% collagenase II (Sigma-Aldrich, St. Louis, MO) at 37 °C for 4 h. Digested chondrocytes were filtered through a 70 µm nylon filter (Becton Dickinson, Franklin Lakes, NJ) and plated into T25 flasks in growth medium which consisted of DMEM with 10% FBS (Gibco, Invitrogen, Carlsbad, CA) as passage 0. The cells were incubated at 37 °C in a humidified atmosphere with 5% CO₂ and medium was replaced twice per week. The cells at passage 3 were utilized for subsequent experiments.

The cytotoxicity of PN-KGN nanospheres on cell viability was evaluated by the Cell Counting Kit-8 (CCK-8) assay (Beyotime, Jiangsu, China). Briefly, passage 3 chondrocytes were plated in 96-well plates at a density of 5000 cells/well and then treated with different amounts of particles that can respectively maximally release KGN to 0, 10, and 100 nM according to the release curve. After incubation for 4 or 7 days with the nanospheres, the absorbance was measured using a spectrophotometric reader (SpectraMax 384, Sunnyvale, CA) at 450 nm. Cell numbers were compared to non-treatment control (n = 3).

The pro-inflammatory activity of PN-KGN nanospheres was evaluated by measuring cytokines secreted from chondrocytes. Briefly, passage 3 chondrocytes were plated in six-well plates at a density of 1 × 10⁵ cells/well and then treated with lipopolysaccharide (LPS; 1 µg/mL), PN, and PN-KGN that can release KGN to 100 nM. The mediums were collected at each time point and concentrations of secreted interleukin-6 (IL-6) in the culture medium were assessed using the enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s instructions (R&D, Minneapolis, MN).

Ten-week-old male SD rats used in the animal experiments were provided by the Animal Research Committee of Zhongshan Hospital, Fudan University. All procedures were carried out according to the guide for the care and use of laboratory animals. OA was surgically induced by anterior cruciate ligament transection (ACLT) and destabilization of the medial meniscus according to previous reports (Hayami et al., 2006). The rats were randomized into four groups: Control group, IA injections of 100 µL saline; PN group, IA injections of 100 µL saline with PN (1.18 mg/mL); KGN group, IA injections of 100 µL saline with 100 µM KGN; and PN-KGN group, IA injections of 100 µL saline with PN-KGN (1.18 mg/mL) that can release 100 µM KGN (n = 6 per group). IA injections were performed every 3 weeks at 0, 3, 6, and 9 weeks. Rats were sacrificed for analysis at 3, 6, and 12 weeks after OA induction (Figure 1(D)).

After euthanasia, the surrounding soft tissues were removed and knee joints were carefully dissected without damaging the cartilage. The cartilage surface was then fully exposed and examined macroscopically. Macroscopic pictures were recorded by a digital camera (M205FA, Leica, Wetzlar, Germany).

The dissected samples were fixed in 4% paraformaldehyde for 1 day and then decalcified in 10% EDTA for 6 weeks. After serial dehydration, the joints were embedded in paraffin and sagittally sectioned at 5 μm thickness. The sections were stained with safranin-O/fast green. The Osteoarthritis Research Society International (OARSI) cartilage histopathology assessment system was used to evaluate the degenerative status and OARSI scores were calculated using the formula: Score = Grade × Stage (Pritzker et al., 2006). Immunohistochemistry was performed to evaluate the metabolism of articular chondrocytes. Anti-type II collagen and anti-type I collagen mouse monoclonal antibodies (Novus, Littleton, CO) were used as primary antibodies (1:100). Detailed procedures of immunohistochemistry are described in our previous study (Fan et al., 2016).

Statistical analyses were performed by GraphPad Prism (GraphPad Software Inc, La Jolla, CA). All data were expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used to assess group differences. p < 0.05 was considered statistically significant from at least three independent experiments.

Successful conjugation of PN-KGN was confirmed by FTIR and ¹H NMR. The FTIR spectrum of PN-KGN showed enhanced peaks at 1703 cm⁻¹ compared to that of PN, indicating the formation of amide bonds. At the same time, there were no absorption bands of ammonium salt (N–H) at 3200–2500 and 2150 cm⁻¹, respectively, indicating that the conjugation between PN and KGN was not the formation of ammonium salt (Figure 2(A)). The ¹ H NMR spectra of PN-KGN showed the major peaks of the benzene rings in KGN at 7.3–7.9 ppm along with enhanced peaks of amide bonds at 7.2 ppm (Figure 2(B)). These results confirmed the conjugation of PN and KGN via the formation of amide bonds.

TEM and DLS were used to characterize the morphology and determine the size of PN-KGN. The particles were spherical and regular in shape with an average size of 25 nm (Figure 2(C)). Figure 2(D) showed the sustained and controlled release of KGN from PN-KGN in vitro.

The cytotoxicity of PN-KGN nanospheres was evaluated in chondrocytes. Figure 3(A) showed the amounts of chondrocytes after being exposed to the different amounts of PN-KGN that could release KGN from 0 to 100 nM. No significances were observed compared to the untreated chondrocytes at 4 and 7 days. The result indicated that both PN and PN-KGN nanoparticles had no cytotoxicity in chondrocytes.

The pro-inflammatory activity of PN-KGN nanospheres was also evaluated in chondrocytes. LPS treatment significantly increased the secretion of inflammatory cytokine IL-6 with time. However, addition of PN or PN-KGN did not increase the secretion of IL-6 compared to the untreated chondrocytes at 24 and 48 h. (Figure 3(B)). The result indicated that PN-KGN nanoparticles had no pro-inflammatory effects on chondrocytes in vitro.

The gross appearance of cartilage was evaluated in each group at 3, 6, and 12 weeks. Although there was no obvious macroscopic cartilage abrasion at 3 weeks in each group, cartilage degeneration characterized by fibrillation, surface erosion, pitting, ulceration, and osteophyte formation, as well as subchondral bone exposure, was observed in Control group and PN group at 6 and 12 weeks; however, the rats in PN-KGN group displayed less extensive cartilage destruction at 12 weeks (Figure 4). The macroscopic results were consistent with the histologic staining. Control group and PN group rats showed progressive cartilage degeneration over time with delamination of superficial layers at 6 weeks and surface denudation at 12 weeks. Cartilage in KGN group rats showed minor surface destabilization and thin cartilage at 6 weeks while the cartilage degeneration became obviously at 12 weeks with matrix vertical fissures. By contrast, Cartilage in PN-KGN group rats showed generally intact surfaces and only mild superficial fibrillation even at 12 weeks (Figure 5(A)). Analysis of the OARSI scores revealed no significant differences between Control group and PN group at each time point. However, these two groups had significantly higher OARSI scores compared with KGN group and PN-KGN group at 6 and 12 weeks. Moreover, the OARSI scores were significantly higher in KGN group than that in PN-KGN group at 12 weeks (Figure 5(B)). The metabolism of cartilage matrix was further verified by immunohistochemistry for collagen II (Col II) and collagen I (Col I). Figure 6 showed typical cartilage erosion as well as subchondral bone exposure in Control group and PN group with weak Col II staining. Cartilages of KGN group displayed clusters of chondrocytes and intense Col I staining. However, cartilages of PN-KGN group displayed much stronger Col II-specific staining and less Col I-specific staining compared to other groups at 12 weeks, indicating IA injection of PN-KGN nanospheres could retain the normal compositions of cartilage matrix for a long time.