Dopamine-triggered one-step functionalization of hollow silica nanospheres for simultaneous lubrication and drug release

Osteoarthritis (OA) has been regarded as a lubrication deficiency related joint disease. Combination of both joint lubrication and drug intervention may provide a promising nonsurgical strategy for treatment of OA. Developing novel and simple approaches to fabricate superlubricating nanoparticles with drug release property is highly required. Herein, dopamine triggered one-step polymerization method was employed to fabricate polydopamine/poly(3-sulfopropyl methacrylate potassium salt) (PDA—PSPMA) conjugate coating on hollow silica (h-SiO2) nanosphere surfaces to engineer functional nanoparticles (h-SiO2/PDA—PSPMA). The as-prepared h-SiO2/PDA—PSPMA exhibits excellent aqueous lubrication performance on biomaterial substrates as well as natural bovine articular cartilage based on hydration effect of negatively charged PDA—PSPMA coating and “rolling” effect of h-SiO2 nanospheres. In vitro drug loading-release experiments demonstrate that PDA—PSPMA coating functionalized h-SiO2 nanospheres show high drug-loading and sustained-release capability of an anti-inflammatory drug, diclofenac sodium (DS). Such h-SiO2/PDA—PSPMA nanospheres can be potentially used as a synergistic therapy agent for OA treatment combining by simultaneous joint lubrication and drug release.


Introduction
Biological aqueous lubrication between sliding articular cartilage surfaces provides an ultralow coefficient of friction (0.001-0.03) under a wide range of conditions. Such superior lubrication remarkably decreases the wear of cartilage over decades of use [1][2][3]. However, a failure of articular lubrication system commonly induces increased wear of cartilage and thus osteoarthritis (OA), which gives rise to severe pain and swelling of joint [4,5]. For early OA, it is often necessary to inject artificial joint synovial fluid as viscous supplements to improve the lubrication situation, such as hyaluronic acid (HA). However, rapid degradation of HA in vivo limits the efficiency of OA treatment and also brings the new risk of external infection caused by frequent injection [6,7]. In addition, supplement of anti-inflammatory drugs can effectively alleviate the progression of OA. Accordingly, a synergetic therapy combining both joint lubrication and drug intervention is considered to be a promising nonsurgical strategy for treatment of OA [8][9][10][11]. Recently, hydration lubrication mechanism and the synergism between different molecular components from the synovial joints (namely hyaluronan, lubricin, and phospholipids) proposed by Klein, have been *Corresponding authors: Qiangbing WEI, E-mail: weiqiangbing@nwnu.edu.cn; Shuanghong MA, E-mail: mashuanhong@licp.cas.cn www.Springer.com/journal/40544 | Friction widely recognized to well explain the superior lubrication of articular cartilage [12][13][14][15]. Large amount of water molecules could tenaciously bind to charged biomacromolecules to form efficient hydration layer, which can serve as a fluid-like boundary film to significantly reduce friction and bear joint pressures. Inspired by hydration lubrication mechanism, researchers have been committed to develop biomimetic superlubricating materials to solve the friction and wear issues in biological systems. Meanwhile, zwitterionic polymers (e.g., phosphatidylcholine headgroups) and various polyelectrolytes were grafted onto biomaterial surfaces as biomimicking systems, which have been proved to be able to significantly reduce friction coefficient [16][17][18][19][20][21]. Furthermore, these hydrophilic polymers functionalized micro-and nanosized particles represent another type of additives for biological aqueous lubrication. The high hydration capacity would make the particles homogeneously disperse in aqueous media and thus enhance lubrication of aqueous fluids [8,[22][23][24][25]. Additionally, the "roll" effect of nanospheres may also contribute to lubrication and anti-wear properties of aqueous fluids. For instance, ball-bearing-inspired polyampholyte-modified hydrogel microspheres were fabricated by microfluidic technique and surface grafting modification. The grafting of zwitterionic brushes enabled microspheres efficient lubrication, reduced degradation, and sustained drug release, which exhibited great potential for the treatment of OA [26].
As for oral anti-inflammatory drugs, it is difficult to reach the joint owing to lack of blood vessels in articular cartilage and small drug molecules can be quickly cleared from the joint. Therefore, encapsulating drug molecules into nanocarriers is an efficient strategy to directly deliver drugs into the joint by local injection [27,28]. Hollow SiO 2 nanoparticles (h-SiO 2 ) with good biocompatibility, stable chemical properties, and the feasibility for surface functionalization, can be a representative drug nanocarrier to achieve controlled drug release [29][30][31]. To improve the dispersibility and endow its good hydration lubrication ability, various hydrophilic polymer brushes were grafted from SiO 2 nanoparticles surfaces by surface-initiated polymerization strategy. Liu et al. [32] grafted negatively charged poly(3-sulfopropyl methacrylate potassium salt) brushes (PSPMA) on the surface of hollow SiO 2 microspheres through surface-initiated atom transfer radical polymerization (SI-ATRP) strategy. The PSPMAfunctionalized h-SiO 2 demonstrated excellent hydration lubrication performance and sustained drug release ability. Furthermore, mesoporous silica nanospheres were also grafted with PSPMA and zwitterionic poly(2methacryloyloxyethyl phosphorylcholine) (PMPC) brushes via photopolymerization [33,34]. The tribology and drug loading-release experiments revealed enhanced lubrication and sustained drug release behavior of the functionalized nanospheres. In addition, the in vitro and in vivo experiments suggested that the functionalized nanospheres not only protected oxidative stress-induced chondrocytes from degradation but also inhibited the development of OA. However, these methods, especially ATRP strategy, inevitably introduces toxic catalysts (e.g., copper ion) or initiators, which have a great impact on the biocompatibility and limit the biological applications of such polymergrafted nanoparticles. Hence, developing novel and simple approaches to fabricate functional nanoparticles with simultaneous lubrication and drug-loading properties for synergetic therapy, but without utilization of complex synthesis and introduction of toxic reagents, is highly required.
Mussel-inspired polydopamine (PDA) surface chemistry has received continually increased attention since the oxidative self-polymerization of dopamine was reported [35]. In particular, PDA coating provides a versatile platform to prepare multifunctional surfaces via post-functionalization and also serves as an adhesion site to assist codeposition with targeted polymers via the non-covalent or covalent interactions [36][37][38][39][40]. However, the polymers used in codeposition process require to be synthesized before modification, which is not a real one-step method. Most recently, Zhang et al. [41] reported a rather simple approach to prepare PDA-polymer conjugate coatings via dopamine triggered one-step polymerization and codeposition of (methyl)acrylate monomers. In this process, dopamine molecules acted as both initiator and adhesive, which greatly simplified the reaction steps and enriched functionality of the coating. After that, dopamine triggered one-step polymerization and codeposition strategy was commonly employed 412 Friction 11(3): 410-424 (2023) | https://mc03.manuscriptcentral.com/friction to fabricate low friction and antifouling coatings on surfaces of macroscale diverse biomaterials [42][43][44][45]. Nevertheless, few studies have investigated the surface functionalization of micro/nanoparticles with dopamine triggered polymerization of monomer, to engineer superlubricating nanoparticles as ideal drug-carrier for treatment of OA.
In our work, the novel and facile approach, dopamine triggered one-step polymerization, was employed to coat h-SiO 2 for fabricating functionalized nanospheres with dual functions of hydration lubrication and drug loading-release properties. Negatively charged 3-sulfopropyl methacrylate potassium salt (SPMA) monomer was used to prepare PDA-PSPMA conjugate coating. As demonstrated in Fig. 1, the h-SiO 2 core is designed as carrier to load and release antiinflammatory model drugs (diclofenac sodium, DS). As for outer PDA-PSPMA coating, PDA matrix endows the coating with considerable surface adhesive ability, while the charged PSPMA segments formed from in-situ polymerization of monomers provide hydration lubrication effect. It is hypothesized that DS loaded h-SiO 2 /PDA-PSPMA could exhibit excellent lubrication in aqueous media based on hydration effect of polymer layer and "rolling" effect of nanospheres. Concurrent with superior lubrication, it also possesses drug loading and sustained release capability. Due to facile and biocompatible fabrication strategy, this functionalized nanosphere can be used as a synergistic therapy agent for OA treatment combining by simultaneous joint lubrication and drug release.

Preparation of hollow SiO 2 nanoparticles
The polystyrene (PS) nanoparticles were fabricated via emulsifier-free emulsion polymerization. St (10.0 g), PVP (1.5 g), AIBA (0.26 g), and H 2 O (100.0 mL) were added into a four-neck round-bottom flask (250 mL). The mixture was bubbled by nitrogen gas for 60 min and then heated to 70 °C for 24 h with mechanical agitation. After polymerization, the dispersion was dialyzed with ethanol to remove impurities to obtain the PS nanoparticles. For the preparation of PS@SiO 2 www.Springer.com/journal/40544 | Friction nanoparticles, PS dispersion (3.5 g) was dissolved into 40 mL of anhydrous ethanol by ultrasonic dispersion, then TEOS (1.0 g) and ammonia (5.0 mL) were added and stirred at 50 °C for 1.5 h. The PS@SiO 2 was collected by centrifugation and washed with water and ethanol. Then the solid was dried under vacuum overnight. PS template was removed by calcining at 100 °C for 1 h, and then 600 °C for 5 h to obtain hollow SiO 2 particles (h-SiO 2 ).

Synthesis of h-SiO 2 /PDA-PSPMA nanoparticles
100 mg of dopamine hydrochloride and SPMA monomer with desired mass were dissolved into 50 mL of Tris buffer solution (10 mM, pH = 8.5). 200 mg of h-SiO 2 nanoparticles were dispersed to the buffer solution under ultrasonication to allow the surface coating of PDA-PSPMA under stirring at ambient temperature for 24 h. After that, the brown-black solution was centrifuged and redispersed with water for three times to obtain black h-SiO 2 /PDA-PSPMA nanoparticles. The resulting h-SiO 2 /PDA-PSPMA sample was dried under vacuum for further characterizations.

Characterizations
The thermal properties were studied on a STA 449 C Jupiter simultaneous TG-DSC instrument (STA449F3, Germany). The element compositions were analyzed using X-ray photoelectron spectra (XPS ESCALAB 250Xi, USA). The UV absorption was recorded on an UV-vis spectrophotometer (Cary 60 UV-Vis, Agilent Technologies, USA). Field emission scanning electron microscopy (FE-SEM Quanta FEG 250, USA) and transmission electron microscopy (TEM, FEI, Tecnai, G2 TF20, USA) were employed to observe the morphology of nanoparticles. The Brunauer-Emmett-Teller (BET) measurements were performed on an ASAP 2020 (Micromeritics Instruments, USA) nitrogen adsorption instrument. The dynamic light scattering technique (DLS) was employed to measure hydrodynamic diameters and ζ-potential of nanoparticles dispersion on a particle size analyzer (Zetasizer Nano ZS, Malvern Instruments, UK) equipped with a 633 nm He-Ne laser.

Evaluation of lubrication performance
The lubrication performance of h-SiO 2 and h-SiO 2 / PDA-PSPMA were evaluated by conventional pinon-disk reciprocating tribometer (CSM Co. Ltd., Switzerland). The PDMS pins with a diameter of 6 mm and untreated silicon wafer (surface roughness: Ra ~ 0.392 nm; Rq ~0.527 nm) were employed as the friction pairs. PDMS pins were prepared from commercially available silicone elastomer kits (SYLGARD 184 silicone Elastomer, base and curing agents, Dow Corning). A polystyrene 96-well cell culture plate with a round bottom was used as a mold to prepare PDMS pins with hemispherical ends. The base and curing agent of the SYLGARD 184 elastomer kit were mixed at a mass ratio of 10:1 and transferred to the mold after removing bubbles under gentle vacuum and then incubated in an oven at 70 °C for 6 h. All the friction tests were done in a reciprocating mode and under different loads and reciprocating frequencies. h-SiO 2 /PDA-PSPMA aqueous suspension with different concentrations were added between the two friction pairs as the lubricants (oscillation amplitude: 5 mm; for a duration of 3,600 sliding cycles). All samples were repeated at least three times to obtain the average value. The natural articular cartilage specimens (size: 1.5 mm × 1.0 mm) used for friction tests were derived from market from the knee joint of a cow, followed by ultrasonic treatment with 75% ethanol and PBS buffer solution for 3 times. The cartilage specimens were fixed onto the machine through commercial glue.

Preparation of calibration curves
Anti-inflammatory drug (diclofenac sodium, DS) was dissolved into PBS buffer solution (pH=7.4) and diluted to different concentrations. The UV absorbance of DS solutions was recorded, and then the calibration curves were drawn based on the different concentration of DS solutions and UV absorbance at 276 nm.

Fabrication of drug-loaded h-SiO 2 /PDA-PSPMA
20 mg h-SiO 2 /PDA-PSPMA and 20 mg DS were added into a beaker containing 20 mL of PBS buffer solution. The mixture was uniformly dispersed under mild ultrasonication and then stirred at 25 °C for 48 h. The obtained suspension was centrifuged at 8,000 rpm for 5 min to separate the precipitates from supernatant. The precipitates were dried under vacuum at 60 °C for 12 h and then weighed to calculate drug loading | https://mc03.manuscriptcentral.com/friction capacity (LC). 1 mL of the supernatant was diluted and then measured by UV-vis spectrophotometer at 276 nm to calculate drug encapsulation efficiency (EE). Each sample was measured three times to ensure accuracy. The LC and EE were evaluated by the following equations, respectively. The same method was used to prepare the drug-loaded h-SiO 2 .

In vitro drug release
To study in vitro drug release behavior, DS-loaded h-SiO 2 /PDA-PSPMA nanoparticles were dispersed in deionized water, and put into the dialysis bag (molecular weight cutoff, 8,000-10,000) at room temperature. Dialysis was carried out with 250 mL of PBS buffer solution (pH = 7.4). 3.0 mL of dialysate was taken out for absorbance test at regular intervals, and then 3.0 mL fresh PBS buffer solution was complemented. Free DS and DS-loaded h-SiO 2 were used as the references. The drug release test of DS-loaded h-SiO 2 was identical with that of DS-loaded h-SiO 2 /PDA-PSPMA. As for drug release test of free DS, a certain amount of DS was dissolved in deionized water and sealed in dialysis bag (molecular weight cutoff, 8,000-10,000). Then the dialysis process was the same with that of DS-loaded h-SiO 2 /PDA-PSPMA. The concentration of DS in the dialysate is calculated by UV-vis absorbance and calibration curve, and the amount of drug release can be calculated by Eq. (3): where M t is the amount of DS released from the as-prepared nanoparticles at time t, while  M is the total amount of DS encapsulated by nanoparticles.

In vitro cell viability assays
The in vitro cytotoxicity of h-SiO 2 /PDA-PSPMA was evaluated via standard Methyl Thiazolyl Tetrazolium (MTT) assay using C57BL/6 Mouse Mesenchymal Stem (MSC) cells. The MSC cells were seeded in a 96-well plate (100 μL/well) and cultured in a standard cell culture medium (containing 5 mL fetal bovine serum, 0.5 mL L-glutamic acid, 0.5 mL double antibody, and 45 mL basal medium) at 37 °C with 5% CO 2 for 24 h. Then h-SiO 2 /PDA-PSPMA dispersions with different concentrations were added and incubated for another 48 h. Finally, the supernatants were removed while cells were washed with PBS solution. The MTT solution was added to each well and incubated at 37 °C with 5% CO 2 for 4 h. Afterwards, the supernatants were removed and 100 μL DMSO (dimethyl sulfoxide) was added to each well. The cells were further incubated for 40 min. The absorbance at 570 nm was recorded by a microplate reader to calculate the cell survival rate. The viability of cells was calculated by Eq. (4): was obtained from only culture medium.

Preparations and characterizations of h-SiO 2 / PDA-PSPMA nanospheres
As shown in Fig. 1, the h-SiO 2 nanospheres were synthesized by the template method. PS nanoparticles were used as templates, which were removed by calcination after h-SiO 2 shell was coated. Then, PDA-PSPMA conjugate coating was deposited on the surface of h-SiO 2 in tris buffer solution containing dopamine molecules and SPMA monomer via dopamine triggered one-step polymerization. During the polymerization, abundant free radicals generated from the oxidative self-polymerization process of dopamine could trigger the polymerization of SPMA monomer. The in-situ generated PSPMA polymers would be incorporated into PDA coating [41]. The surface chemical compositions of h-SiO 2 and h-SiO 2 /PDA-PSPMA were evaluated by X-ray photoelectron spectroscopy (XPS). As shown in Fig. 2(a), www.Springer.com/journal/40544 | Friction the binding energies of Si 2p and Si 2s were at 103 and 153 eV, respectively, which was attributed to the silicon element in h-SiO 2 [46,47]. After deposition of PDA-PSPMA coating, the signals of Si 2p and Si 2s were obviously decreased, indicating SiO 2 surface was covered with polymer coating. The successful deposition of PDA-PSPMA coating was further confirmed by the appearance of N 1s signal at 400 eV and S 2p signal at 168 eV, which were not found in bare h-SiO 2 [33]. To estimate the mass fraction of polymer in h-SiO 2 /PDA-PSPMA hybrids, thermogravimetric analyses (TGA) of h-SiO 2 and h-SiO 2 /PDA-PSPMA were conducted. As shown in Fig. 2(d), a negligible weight loss of 1.9% was observed from the curve of h-SiO 2 , demonstrating the PS template was totally removed. When the PDA-PSPMA coating was deposited, the weight loss was increased to 14.7%. As a result, the mass fraction of the PDA-PSPMA coating is calculated to be 12.8%. The TGA data not only confirm the successful deposition of polymer coating on h-SiO 2 surfaces, but also provide a quantitative assessment for each component of the nanohybrids.
The morphologies of PS, h-SiO 2 , and h-SiO 2 /PDA-PSPMA were observed via SEM and TEM. As shown in Figs. 3(a) and 3(d), PS templates showed well-defined spherical morphology and uniform size distribution with the average diameter of ca. 290 nm. The surface of PS nanosphere was quite smooth. PS@h-SiO 2 was prepared by the sol-gel process. After the removal of PS templates by calcination, the high contrast between dark shells and light cavities suggested that PS templates were totally removed and hollow structure was formed. The surface of h-SiO 2 nanosphere became rougher and the thickness of h-SiO 2 shell was about 14 nm [32,48,49]. Figures 3(c) and 3(f) demonstrate the SEM and TEM images of h-SiO 2 /PDA-PSPMA nanospheres, respectively. After coated with PDA-PSPMA, the morphology of h-SiO 2 was well maintained and the outer shell became less uniform and much rougher. Thickness of the shell layer was measured to be ca. 20 nm, indicating that the thickness of the PDA-PSPMA deposited layer was approximately 6 nm. TEM images also confirm that h-SiO 2 was successfully coated with PDA-PSPMA.  To further study the porosity of h-SiO 2 and h-SiO 2 / PDA-PSPMA, N 2 adsorption-desorption isotherms and corresponding pore size distribution curves of h-SiO 2 and h-SiO 2 /PDA-PSPMA were measured. As shown in Fig. 4, both isotherms were in accordance with the typical IV type N 2 adsorption/desorption pattern [46], indicating that the nanoparticles have a mesoporous structure [49]. The specific surface area of h-SiO 2 was calculated to be 40.9 m 2 /g according to the Brunauer-Emmett-Teller (BET) model, while the pore size was about 3.8 nm with relatively narrow distribution. Following the deposition of PDA-PSPMA coating, the specific surface area and pore size of the h-SiO 2 /PDA-PSPMA were reduced to about 29.2 m 2 /g, and about 3.1 nm, respectively. As for h-SiO 2 /PDA-PSPMA, the decrease of specific surface area and pore size was attributed to the coverage of PDA-PSPMA polymer layer on h-SiO 2 surface. The relatively large specific surface area and cavity structure of h-SiO 2 / PDA-PSPMA nanospheres provided enough space for drug loading, and the pores on the surface provide channels for drug loading and release [46,48].   The hydrodynamic diameter and ζ-potential of nanospheres in aqueous media were evaluated using dynamic light scattering (DLS). Figure 5 shows the DLS intensity distribution of h-SiO 2 and h-SiO 2 /PDA-PSPMA at 25 °C in aqueous solution. The corresponding average diameters and ζ-potentials were summarized in Table 1. The hydrodynamic diameter of h-SiO 2 was about 373.2 nm, which was a little bit larger than the average diameter of h-SiO 2 in dry state (317.1 nm). This is due to adsorbed hydration layer on h-SiO 2 surface. The hydrodynamic diameter of h-SiO 2 /PDA-PSPMA was around 471.8 nm, which was much larger than that of h-SiO 2 . Furthermore, as shown in Table 1, after coated with PDA-PSPMA, the ζ-potential of the nanospheres changed from nearly neutral to negative (ca. -7.8 mV). These results further verified that negatively charged PDA-PSPMA coating was successfully deposited on h-SiO 2 surfaces. It was worth noting that the hydrodynamic diameter of h-SiO 2 / PDA-PSPMA was much larger than that in in dry state (328.2 nm) tested by TEM. This is due to high hydration capacity of negatively charged PSPMA. PDA-PSPMA conjugate coating could swell and form a hydration layer by combining large amount of water molecules [50]. Therefore, DLS analysis demonstrated that the PDA-PSPMA coated h-SiO 2 dispersions with high hydration have a great potential for aqueous lubrication [51].

Evaluation of lubrication performance
The lubrication performance of h-SiO 2 /PDA-PSPMA aqueous dispersion was evaluated on a ball-on-disk reciprocating tribometer using silicon wafers and PDMS soft balls as friction pairs. The coefficients of friction (COF) of pure water, bare h-SiO 2 as well as PDA coated h-SiO 2 (h-SiO 2 /PDA) were also presented as comparisons. Figure 6(a) shows the COF vs time plots of different samples under the load of 0.5 N and with the reciprocating frequency of 1 Hz. The high/non-stable COF of water can be attributed to its low viscosity and large elastic deformation along with rising wear of PDMS pair. Compared with pure water, bare h-SiO 2 dispersion (2 mg/mL) exhibited obvious frictionreduction property with average COF of ~0.3, possibly due to "rolling" effect of spherical nanoparticles. However, the COF becomes non-stable and increased obviously after 2,200 s, the reason for this may be that the hollow nanoparticles were broken under dynamic shearing process. In contrast, h-SiO 2 /PDA-PSPMA dispersion displayed ultralow COF after 200 s of "running in" process and could keep in constant even   Fig. 6(b). With increasing the concentration of h-SiO 2 /PDA-PSPMA dispersion from 1 to 2 mg/mL, the mean COF decreased from 0.22 to 0.04. While continually increasing the concentration of dispersion (from 2 to 4 and 6 mg/mL), the mean COF showed an increasing trend, from 0.04 to 0.12 and then 0.21. These results illustrated that high concentrations of h-SiO 2 /PDA-PSPMA would lead to relatively poor lubricating properties. This is because high concentration dispersion may result in agglomeration of nanoparticle and thus poor lubrication. According to previous reports, acrylate monomers in the system can inhibit the formation of PDA nanoaggregates and then decrease adhesion and deposition of PDA species on solid surface [52]. Therefore, the effect of mass ratios of dopamine/SPMA (content of SPMA monomer) on the tribological properties of h-SiO 2 /PDA-PSPMA were investigated (loading force: 0.5 N; reciprocating frequency: 1 Hz; and concentration: 2 mg/mL). As shown in Fig. 6(c), the average COF decreased from 0.15 to 0.07 and then 0.04 with increasing the content of SPMA monomer (the corresponding dopamine/SPMA mass ratios were 1:5, 1:10, and 1:20, respectively). h-SiO 2 /PDA-PSPMA prepared with dopamine/SPMA mass ratio of 1:20 exhibited best lubrication. When dopamine/SPMA mass ratio is 1:30, the COF was slightly increased to 0.05. This phenomenon is resulted from the fact that high concentration of SPMA can inhibit the formation of PDA nanoaggregates and then influence the deposition of PDA on h-SiO 2 surface. Afterward, the lubrication performance of h-SiO 2 /PDA-PSPMA under different loading force and at different reciprocating frequency were evaluated. With increasing the normal loads from 0.5 to 3 N, the COF gradually increased (Fig. 6(d)), which was attributed to the deformation of soft PDMS as well as partially damage of PDA-PSPMA hydration layer surrounding h-SiO 2 nanospheres. In Fig. 6(e), the COF presented a decreasing trend with increasing the reciprocating frequency from 0.5 to 1 Hz, while it began to increase when reciprocating frequency was larger than 1 Hz. As h-SiO 2 /PDA-PSPMA was intended to reduce friction and load drugs for sustained release in biological media. It is necessary to study the lubrication performance of drug-loaded h-SiO 2 / PDA-PSPMA (DS-h-SiO 2 /PDA-PSPMA) and h-SiO 2 / PDA-PSPMA in saline solution (0.9% NaCl) and PBS buffer (pH = 7.4). As shown in Fig. 6(f), DS-h-SiO 2 / PDA-PSPMA also exhibited excellent lubrication in water with average COF of 0.03, indicating that loading of DS drug doesn't influence the superior lubrication of h-SiO 2 /PDA-PSPMA. The average COFs of h-SiO 2 / PDA-PSPMA were approximately 0.06 and 0.10 in PBS buffer and saline solution, respectively. Comparing with the COF in pure water, h-SiO 2 /PDA-PSPMA could maintain the superior lubrication in PBS buffer, whereas the lubrication capability was slightly weakened in saline solution. This slight increase of COF was possibly attributed to the partially dehydration of PSPMA chains in the coating arising from electrostatic screening effect of electrolytes. But h-SiO 2 /PDA-PSPMA still showed much better lubrication than h-SiO 2 /PDA and h-SiO 2 even though the slight increase of COF in saline.
To examine the feasibility of h-SiO 2 /PDA-PSPMA dispersion on different substrates, the tribological tests were also performed on titanium alloy (Ti 6 Al 4 V), stainless steel (316L), glass, and even articular cartilage. As shown in Fig. 7(a), it was clear that h-SiO 2 /PDA-PSPMA water dispersion presented very good lubrication properties on commonly used biomaterials surfaces. The slight increase in COFs on Ti 6 Al 4 V and 316L may be caused by relatively large roughness of the substrates. Particularly, h-SiO 2 /PDA-PSPMA showed very efficient lubrication on articular cartilage with low COF of ca. 0.022, demonstrating its promising potential as injectable lubricants for restoring the lubrication of joints with loss of lubrication function. Figure 6(b) showed the comparison of lubrication capability of as-prepared h-SiO 2 /PDA-PSPMA with commercially available biolubricants, hyaluronic acid (HA), and bovine serum albumin (BSA). The articular cartilages were used as friction pairs for both two sides. It could be seen that h-SiO 2 /PDA-PSPMA exhibited much better lubrication than BSA aqueous solution. Furthermore, the friction curve of h-SiO 2 / PDA-PSPMA without "running in" process showed similar trend with that of HA, suggesting the efficient lubrication capability of these two lubricants. However, the average COF of h-SiO 2 /PDA-PSPMA (0.021) was lower than that of HA (0.024) on articular cartilage surfaces, indicating that h-SiO 2 /PDA-PSPMA was more efficient than commercial HA for lubrication of degraded cartilage.
Based on the friction results of h-SiO 2 /PDA-PSPMA, a plausible lubrication mechanism of h-SiO 2 /PDA-PSPMA nanoparticles was proposed and given in | https://mc03.manuscriptcentral.com/friction was achieved by the combination of hydration lubrication and rolling lubrication effect [53]. The negatively charged PDA-PSPMA coating on surface of h-SiO 2 nanoparticles could form efficient hydration layer for achieving good hydration lubrication [54]. Meanwhile, hydration layer formed by charged SO 3 groups and water molecules could bear normal pressure. At the same time, the bounded water molecules exchange with free water molecules fast to give a fluid-like interface. Such fluid-like interface surrounding the nanoparticles results in ultralow COF under shearing condition. Additionally, the hydration layer could prevent h-SiO 2 /PDA-PSPMA nanoparticles from aggregation. Thus, it can be speculated that the free rolling of h-SiO 2 /PDA-PSPMA nanoparticles may employ a rolling mechanism to reduce friction between sliding surfaces. In this case, the soft and fluid-like PDA-PSPMA hydration layer and the hard h-SiO 2 nanoparticles play a synergistic effect and co-contribute to the superior lubrication of h-SiO 2 /PDA-PSPMA water dispersion.

In vitro drug loading and release
As h-SiO 2 /PDA-PSPMA nanospheres could achieve a ultralow COF in aqueous media as well as biological fluids, which is expected to be a promising biological lubricant for treatment of OA. In addition, the hollow structure and rich pores on the surface make h-SiO 2 / PDA-PSPMA nanospheres as an ideal drug carrier for drug loading and release. The h-SiO 2 /PDA-PSPMA nanospheres can realize the dual function of lubrication and drug loading. Herein, DS, an anti-inflammatory and analgesic drug for OA treatment, was used as a model drug to evaluate sustained release property of h-SiO 2 /PDA-PSPMA. The calibration curve and chemical structure of DS were presented in Fig. 9 [55]. These results demonstrated that h-SiO 2 /PDA-PSPMA possessed good drug loading capacity and sustained release performance.

In vitro cytotoxicity
Nontoxicity or low toxicity is a critical parameter for biomedical applications of h-SiO 2 /PDA-PSPMA nanospheres. Therefore, in vitro cytotoxicity was investigated by MTT assay. MSC cells were incubated with different concentration of h-SiO 2 /PDA-PSPMA nanospheres for 48 h. As shown in Fig. 10, compared

Conclusions
In this work, the dopamine-triggered one-step surface functionalization method was introduced to coat h-SiO 2 with PDA-PSPMA conjugate coating for developing dual functional nanospheres (h-SiO 2 /PDA-PSPMA) with simultaneous aqueous lubrication and drug release properties. A series of characterizations were employed to confirm the successful fabrication of h-SiO 2 /PDA-PSPMA. The lubrication tests results reveal that obtained h-SiO 2 /PDA-PSPMA nanospheres exhibit superior aqueous lubrication performance on biomaterial substrates as well as natural bovine articular cartilage, mainly attributing to hydration mechanism of negatively charged PDA-PSPMA coating and free "rolling effect" of spherical h-SiO 2 nanoparticles. Moreover, in vitro drug release demonstrates that h-SiO 2 /PDA-PSPMA possesses a good drug loading and sustained release abilities of an anti-inflammatory drug (e.g., DS). Due to facile fabrication strategy, efficient lubrication, drug loading and release behavior, as well as good biocompatibility,