Intra-articular Administrated Hydrogels of Hyaluronic Acid Hybridized with Triptolide/Gold Nanoparticles for Targeted Delivery to Rheumatoid Arthritis Combined with Photothermal-chemo Therapy


 Triptolide (TP) as a disease-modifying anti-rheumatic drug (DMARD) is effective on the treatment of rheumatoid arthritis (RA). To alleviate the toxicity and elevate therapeutic specificity, hyaluronic acid (HA) hydrogels load RGD-attached gold nanoshell containing TP are synthesized, which can be used for targeted photothermal-chemo therapy, and imaging of RA in vivo. The hydrogels system composed of thiol and tyramine modified HA conjugates has been applied artificial tissue models of cartilage for studying drug delivery and release properties. After the degradation of HA chains, heat together with drugs can be delivered to the inflammatory joints simultaneously due to the near-infrared resonance (NIR) irradiation of Au nanoshell. RA is a chronic inflamed disease, which is characterized by synovial inflammation of multiple joints, and can be penetrated with NIR light. These intra-articular administrated hybrid hydrogels combined with NIR irradiation can improve clinical arthritic conditions and inflamed joints in collagen-induced arthritis (CIA) mice, which just need a smaller dosage of TP with non-toxicity. Additionally, the TP-Au/HA hybrid hydrogels treatment reduced the invasion and migration of RA fibroblast-like synoviocytes (RA-FLSs) in vitro significantly, through reducing the phosphorylation of mTOR and p70S6K, its substrates, and confirmed that the mTOR pathway was inhibited.


Introduction
Rheumatoid arthritis (RA) is a systematic in amed disease, which is characterized by aggressive joint destruction and persistent synovitis [1]. The main treatment of RA is to use disease-modifying antirheumatic drugs (DMARDs), which can improve in ammatory conditions and delay progression [2]. As an epoxide diterpene lactone compound, TP has been used clinically to inhibit the expression of adhesion molecules, metalloproteinases, and pro-in ammatory cytokine, thus demonstrating anti-in ammatory and cartilage protective effects in the treatment of RA [3][4][5]. However, as a DMARD, due to its multiple organ toxicity, less solubility, low bioavailability, less stability, rapid excretion and non-speci c targeting, its anti-arthritic e cacy was not fully exerted in clinical [6,7]. Therefore, the novel delivery systems will be an effective tool for RA treatment, which maintain a higher concentration of TP in the main in amed area of joints, and smaller-dosage administration is used to reduce side effects. lubrication, provides a renewed source of HA for joint tissues, cushions the load transfer across articular surfaces and anti-in ammatory properties for synovial uid [6,16].
Through physically crosslinked or chemically modi ed improvement, HA based scaffolds, mechanically soft/tough hydrogels, in-situ hydrogels and viscoelastic solutions have been prepared successfully [6,12]. However, most hydrogels may lead to the uncontrolled and rapid release of tiny bare molecules because of the high-water substance and large size of pore [17]. Particularly, hydrophobic drugs with less water-soluble lead to simply release precipitate or from the hydrogels rapidly [17]. Hybrid hydrogels has been synthesized to overcome the limitations, the delivery systems is the combination of drug-loaded nanoparticles with hydrogels [18]. Among various kinds of Au nanostructures, nanoshells, nanocages and nanorods, have attracted a potential interest in the various elds such as bio-sensors, drug delivery, imaging, tissue engineering and photothermal therapy [19][20][21]. Au nanostructures have good stability and biocompatibility, excellent properties of optics and electricity, and the easiness of surface modi cation in vivo [19,22]. Meanwhile, Au nanostructures exhibit the unique surface plasmon resonance and intense absorption in NIR region, and due to its strong NIR absorption and localized cytotoxic heat upon NIR irradiation, allowing their use NIR absorbance imaging and photothermal therapy in vivo [19]. In vitro release studies have demonstrated that drug from the metallic nanoparticles hybrid higher cross-linked hydrogels have longer-sustained delivery [17]. These results show that drug delivery systems can be designed to provide a customized release rates and sustained delivery of drugs with appropriately designing of nanoparticles carriers and network structure.
For the treatment of RA, we developed TP-Au/HA hybrid hydrogels with RGD targeting, are directly intraarticular injected into CIA mice, which may potentially minimize side effects [9]. After the degradation of HA chains, TP-Au-RGD nanoparticles are exposed to the surrounding environment [23]. The accumulation of synthesized nanoparticles is enhanced in the in ammatory joints due to the TP-Au nanoparticles contain RGD peptides. Due to Au nanoshell, heat is generated and drugs are released from Poly (DL-lactic-co-glycolic acid) (PLGA)-TP nanoparticles rapidly upon NIR exposure, allow photothermally to control drug delivery and release. The application of hybrid hydrogels earns superior therapeutic effect compared to conventional treatment in CIA mice.
Furthermore, in vivo and in vitro experiments are applied to deeply verify the antirheumatic mechanisms of TP-Au/HA hybrid hydrogels. These synthesized hydrogels are speci cally targeted to in amed joints, and exert superior anti-arthritic effects on reducing destructive joint in ammation mediated by RA-FLSs with low toxicity [24,25]. To reveal the mechanism of targeted drug delivery in in ammatory joints, RA-FLSs are applied for in-depth signaling pathway investigation, which for their characteristic pathogenic behaviors, such as mediate in ammation, aggressive phenotype and destruction of the joint [26]. It indicates that the anti-in ammatory effects of TP-Au/HA hybrid hydrogels on RA-FLSs are occurred by regulating the mTOR/p70S6K signal pathway. mTOR/p70S6K signal pathway is kept in an abnormally activated state, leading to the high expression of anti-apoptotic genes of downstream and then a subsequent in uence on multiple effector molecules of downstream [27,28]. Therefore, the study aims to study the anti-in ammation effects of TP-Au/HA hybrid hydrogels in CIA mice and to reveal the signaling pathways that may be involved.

Fabrication of TP-Au-RGD nanoparticles
Under magnetic stirring, 300 mg of PLGA and 18 mg of TP were dissolved in 30 ml of dichloroethane, and slowly added drop-wise 300 ml of distilled water containing Pluronic F-127 (300 mg) as a stabilizer to the above solution. The mixture was emulsi ed by ultrasonication for 1 h after mixing the oil and water phase, then following evaporation of dichloroethane with stirring for 24h. Then, the TP-PLGA nanoparticles were collected by centrifugation, and then re-dispersed in 10 ml of PBS through sonication. After that, the TP-PLGA nanoparticles solution was added to 50 mL Au nanoparticles (AuNps) solution with vigorously stirred for 20 h. Then, the TP-PLGA-AuNps nanoparticles were collected by centrifugation.
The sodium citrate reduction method is used to prepare AuNps [19]. 64 ml of 0.1 g/L HAuCl 4 solution is heated to boiling. Then the solution stirred strongly at around 120rpm, meanwhile, 0.42 ml of 10 g/L sodium citrate are added dropwise into the solution to keep the reduction time about 6min. After that, the solution is kept boiling until the solution turns red-purple and transferred into a ask and then stored at 4°C before use. Then, 0.315 mL sodium citrate solution and 22.592 mL of HAuCl 4 solution were added to 60 mL TP-PLGA-AuNps under ultrasonic stirring. Afterwards, hydroxylamine solution was added drop by drop and the mixed solution was stirred for 30 min to realize the reduction of HAuCl 4 under the ultrasonication. The formation of Au nanoshell outside the TP-PLGA nanoparticles was by the reducing HAuCl 4 around TP-PLGA-AuNps [19].
The TP-PLGA-Au nanoparticles were dispersed into SH-PEG-COOH solution by sonication from the substrate and collected by centrifugation at 10000 rpm. Under magnetic stirring, the collected carboxylate-terminated TP-PLGA-Au nanoparticles, cyclic RGD (6 mg), 8 mg of NHS and 8 mg of EDC were dispersed in 18 mL of PBS (0.2 M, pH 7.4) at room temperature. During this period, the RGD peptides is covalently bound to the -COOH group of the SH-PEG-COOH chains adsorbed to Au nanoshell. TP-Au-RGD nanoparticles were collected by centrifugation after 24 h, discarded the supernatant that contained unreacted cyclic RGD. Finally, the obtained TP-Au-RGD nanoparticles were freeze-dried and stored.

Synthesis of thiol and tyramine modi ed HA
Synthesis of thiol and tyramine modi ed HA as show in Fig. 2a. HA (0.5 g) was dissolved in 100 ml of deionized water. 0.958 g of EDC and 0.575 g of NHS (5 mmol) were released in deionized water (10 ml), and add to the above HA solution, respectively. The mixture was adjusted pH to 5.4 with 1 M HCl and then stirred 0.5 h. After that, 1.74 g of TA (10 mmol) and 2.25 g of cystamine dihydrochloride (10 mmol) were added to the mixture and then stirred 24 h. 2.3 g of DTT (15 mmol) was added and stirred 24 h. Transferred the mixture solution into a dialyzed for 3 days to ensure the remaining tyramine hydrochloride and other salts completely removed. The cut off molecular weight of dialysis bag is 3.5 k Da. and then. Finally, the obtained -SH and tyramine modi ed HA was freeze-dried.

Preparation of TP-Au/HA hybrid hydrogels
The preparation of the hybrid hydrogel refers to the previous work [29]. -SH and tyramine modi ed HA (0.5 g) was dissolved in 10 ml of PBS at room temperature. After completely dissolved, gently mixed 10 mg of TP-Au-RGD nanoparticles to the prepared solutions and ultrasonication for 10 min. Subsequently, HRP was dissolved in PBS (50 µl, 0.02 mg/ml)was added to the previous mixture rst, then hydrogen peroxide (H 2 O 2 ) (50 µl, 0.02wt %). The mixture was then stirred gently, which resulted in hydrogels formation. The obtained TP-Au/HA hybrid hydrogels was freeze-dried.

Material characterization
The obtained TP-Au-RGD nanoparticles were characterized by a 1 HNMR spectrometer (Varian Gemini-300, DMSO-d 6 ). The zeta-potential and size of obtained TP-Au-RGD nanoparticles were measured by dynamic light scattering at 25°C. The TP-Au-RGD nanoparticles exhibited an average size of 140.3 nm. The zeta potential was found to be − 15.2 ± 0.4 mV. The TP-Au-RGD nanoparticles using TEM was shown is Fig.  2b. The encapsulation e ciency of the prepared TP-Au-RGD nanoparticles were found to be 30 ± 5%.

In vitro release
Initially, 10 mg of TP-Au/HA hybrid hydrogels and equivalent TP-Au-RGD nanoparticles were loaded into two separate dialysis bag, the cut off molecular weight of dialysis bag is 1k Da. The two separate dialysis bag were immersed in a small glass containing with 10 mL of PBS and slight constant shaking at150 rpm, respectively. And NIR irradiation of 0.53 W/cm 2 for 10 min performed at beginning. Due to maintain release conditions, using fresh PBS to replace the release medium at determined intervals to at 37 ℃. Using UV-vis spectrophotometry to measure the amount of released TP at 220 nm.

RA-FLSs preparation and culturing
RA-FLSs and growth medium were purchased from Cell Applications (Beijing Longyue Biological Technology Development Co., Ltd.). RA-FLS cells obtained from passages 5 to 9 were seeded onto 96well plates at a density of 1 x 10 4 cells /mL. They were cultured in Dulbecco's modi ed Eagle medium (DMEM) with 4.5 g/L glucose, 100 µg/mL streptomycin, 100 IU/ml penicillin, and 10% fetal bovine serum. Cells were grown in a humidi ed 37 ℃ incubator with oxygen and 5% CO 2 .

CCK-8 assay
A CCK-8 assay was used to measure the effect of TP-Au/HA hybrid hydrogels on the proliferation of RA-FLSs. In brief, RA-FLSs were collected and seeded onto 96-well plates at a density of 2 × 10 4 cells/well. Then, the cells were plates with different concentration of drug and for different lengths of time (24 h, 48 h). Cells were treated with TP solution (0.13 and 30µM), TP-Au/HA hybrid hydrogels with equivalent TP of o.13µM without NIR irradiation, and TP-Au/HA hybrid hydrogels with equivalent TP of o.13µM with NIR irradiation (0.38 W/cm 2 for 10 min). After the treatments, 10 µl CCK-8 solution (Shanghai yuanye Bio-Technology Co., Ltd.) was added to each well and incubated at 37 ℃, and 5% CO 2 for 1 h. The absorbance was determined at 450 nm by using an enzyme-linked immunosorbent assay reader (BioTek Instruments, Inc., Winooski, VT, USA).

Western blot analysis
mTOR p-mTOR p70S6Kand p-p70S6K protein (Abcam, UK) expression was evaluated by western blotting. Extracts of total protein cell were prepared by the Cell Lysis Buffer (Abcam, UK). Each sample was loaded thirty micrograms of total protein and separated on a 4-20% SDS-PAGE gel under reducing conditions. Then, transferred the samples onto a nitrocellulose membrane, and then blocked in 5% nonfat dry milk. The membranes were incubated with primary antibodies and secondary antibodies. Detection was performed using chemiluminescence.

Induction and treatment of CIA mice
Firstly, bovine type II collagen (200µg, Sigma-Aldrich, Shanghai) emulsi ed in complete Freund's adjuvant (200µg, Sigma-Aldrich, Shanghai) was intravenous injection male DBA/1J mice (8 weeks old, Medcona, China) to induce rheumatoid arthritis, which using an intradermal injection. Next, 100µg of bovine type II collagen in incomplete Freund's (Sigma-Aldrich, Shanghai) was given at 21 days after the primary immunization, which using a booster intradermal injection.
After rheumatoid arthritis was fully developed, saline (G1), TP solution (G2), TP-Au/HA hybrid hydrogels (G3 and 4) were intraarticular administered to the mice, and after TP-Au/HA hybrid hydrogels injection, G4 combined with10 min 1.59 W/cm 2 NIR light (n = 5 mice each group). Mice were monitored twice a week for 4 weeks after intra-articular injection. The clinical index is the amount of the clinical scores for fourpaw clinical scores, the highest score is 16 points [30]. The evaluated paws were scored from 0 to 4 according to the following scale: 0 = no erythema and swelling, 1 = erythema and mild swelling, 2 = erythema and mild swelling extending from the tarsals to the ankle, 3 = erythema and moderate swelling extending from the metatarsal to ankle joints, and 4 = erythema and severe swelling of the digits, foot and ankle or ankylosis of the limb.

In vivo NIR imaging
TP-Au/HA hybrid hydrogels were intraarticular injected into the CIA mice (200 µ L, 1mg/ml dispersed in PBS). The mice were anesthetized with a short acting anesthetic, and kept during the imaging process. IVIS Spectrum (Carestream Health Fx Pro/FX) in vivo uorescence imaging system was used.

Histological examination
The mice were sacri ced after 28 days of each treatment. Removed the joints from the mice, and xed in 10% buffered formalin saline at 4°C for 1 week for histopathological examination. The decalci ed joints were embedded in para n blocks and 4µm-thick para n sections were sliced. Using hematoxylin and eosin (H&E) to stain the joint tissues sections, and score the changes in synovial in ammation on a scale of 0-4 [33]. Each score was assessed by two independent experimenters and the average grades were calculated.

Microcomputed Tomography
Experimental mice paws of each groups were scanned by micro-CT system (SkyScan 1176, SkyScan, Aartselaar, Belgium). Images were acquired at 80 k Vp, 5 s/frame, and 150 mA, with 360 views. The estimated radiation dose was approximately 6.9 mGy using image acquisition protocol. Using NFR Polarys software (Exxim Computing Corporation, Pleasanton, USA) to reconstruct and evaluate the threedimensional structure of scanned paws., And using Aquarius software (version 4.4.6, TeraRecon, Inc.) to measure three-dimensional bone volume (BV) including phalanges and metatarsal bones to con rm volumetric change of arthritis joints.

Biodistribution and clearance
TP-Au/HA hybrid hydrogelswere administered intraarticularly into CIA mice (200µL, 1mg/ml dispersed in PBS) (n = 5).The mice were sacri ced after 28 days of each treatment, and removed the major organs (liver, heart, spleen, kidney and lung) from the mice.

Statistical analyses
All images and data were from three independent experiments. Data are expressed as means ± standard deviation. Statistical analyses of group were performed using the Prism graph pad 8.0, and post-test was followed by Tukey's method.

Preparation and characterization of TP-Au/HA hybrid hydrogels
The preparation route is shown in Fig. 1. First, we prepared TP-PLGA nanoparticles. Next, through the electrostatic interaction, TP-PLGA nanoparticles were modi ed with AuNps orderly to obtain TP-PLGA-AuNps nanoparticles, and HAuCl 4 is reduced around TP-PLGA-AuNps to form Au nanoshell outside the TP-PLGA nanoparticles [19]. AuNps were prepared in boiling condition by reducing and stabilizing HAuCl 4 with sodium citrate [30]. And the TP-PLGA-Au nanoparticles were dispersed into 1 wt % SH-PEG-COOH solution by sonication, and then collected by centrifugation. The cyclic RGD peptide for targeted delivery can bind R v β 3 expressed on angiogenic vascular endothelial cells at in ammatory sites [31]. The size of bare TP-Au-RGD nanoparticles was 140.3 nm in diameter. The zeta potential was found to be − 15.2 ± 0.4. 1 H NMR spectra of the obtained TP-Au-RGD nanoparticles was examined in Fig. 3b, peaks at 8. 27-8.438 ppm correspond to the -NH(NH 2 ) = NH of cyclic RGD and 4.85-5.02 ppm correspond to the CH of PLGA.
The absorption peak of AuNPs at 680 nm to 740 nm in the UV-vis-NIR spectra (Fig. 3c), indicating that Au nanoshell was formatted and nanoparticles can be used for NIR absorbance imaging and photothermal therapy in vivo.
For the preparation of the modi ed HA, hydroxylic groups of hyaluronic acid were activated by HCl and then reacted with tyramine and cystamine [10]. This process resulting in the formation of a carbamate bond between the amine group of HA and the hydroxyl groups, and end-group thiolated HA was synthesized via reductive amination, as shown in Fig. 2a. 1 H NMR spectra of the modi ed HA was showed in Fig. 3a. The new appeared peaks at 6.64 and 6.60 ppm corresponded to the aromatic protons of TA being integrated, and peaks of two methylene groups on cysteamine were at 2.05 and 2.67 ppm.
Finally, through the catalysis of H 2 O 2 and HRP, tyramines and thiol oxidatively coupled to form the hydrogels. The SEM images of modi ed HA showed the surface morphology is smooth in Fig. 2c. The TEM image was evident that TP-Au-RGD nanoparticles were homogenous distributed in the hydrogels in Fig. 2d, and the SEM image of hydrogels in Fig. 2e

Drug release and photothermal effects
In vitro release studies, we measured the release of TP-Au/HA hybrid hydrogels and TP-Au-RGD nanoparticles, which demonstrated longer sustained delivery of drug from the higher cross-linking hydrogels. A relatively dense and stable hydrophilic shell was formed by the long HA chain with multiple interaction sites, as shown in Fig. 1a. TP-Au-RGD nanoparticles showed that the release time of TP is 3 days, and the burst release is more than 60% within 12 h (Fig. 4c). The release pro le of drug from TP-Au/HA hybrid hydrogels showed the sustained release time of TP is 3 days, and the burst release of about 60% of the drug within 72h, which may be related to the embedding of nanoparticles within hydrogels network. The release from TP-Au/HA hybrid hydrogels was slower, which might be related to the combined e cacy of TP-Au-RGD nanoparticles release and coacervate system [32]. After the degradation of HA chains, the surface of the TP-Au-RGD nanoparticles was exposed to the surrounding environment [22,29]. Because the Au nanoshell has absorption in the NIR region, NIR irradiation will locally generate heat which is not enough to cause irreversible tissue damage [33]. And with increasing temperature, the degradation of biodegradable PLGA nanoparticles was more rapidly [34,35]. Then, we measured the release rate of TP at 37°C to study the e cacy of NIR irradiation and non-NIR irradiation (Fig. 4d). The release pro le of TP from TP-Au-RGD nanoparticles with non-NIR irradiation is in a linear, showing the release rate of TP was nearly constant. However, the release rate was reduced and induced a burst release of TP for 12h with 10 min NIR irradiation. These results indicate that NIR irradiation can control the TP release rate from TP-Au-RGD nanoparticles.

The proliferation inhibition effect on RA-FLSs in vitro study
In vitro study, we used RA-FLSs to con rm photothermal controlled release of drugs and antiin ammatory effects of the TP-Au/HA hybrid hydrogels combined with NIR irradiation. CCK-8 assay demonstrated that the proliferation of RA-FLSs was signi cantly inhibited by TP-Au/HA hybrid hydrogels (Fig. 5d). G1 was a control group. And G2 and G3 were treated with 0.13µM and 30µM TP solutions, respectively. G4 and G5 were prepared by culturing cells with TP-Au/HA hybrid hydrogels (equivalent TP of G2) for 1 day. And G5 was also treated with 10 min NIR irradiation at 0.38 W/cm 2 . As expected, due to the lower TP concentration, the cell anti-proliferation effect of G2 was much less than G3. The treatment of TP-Au/HA hybrid hydrogels signi cantly enhance cell anti-proliferation without NIR irradiation (G4). The greatest anti-proliferation effect of cell was the treatment of TP-Au/HA hybrid hydrogels combined with NIR irradiation (G5), although the dosage of TP in G5 was lower than G3. These results demonstrated that TP-Au/HA hybrid hydrogels had a synergistic effect with NIR irradiation.

Effects of TP-Au/HA hybrid hydrogels on the mTOR/p70S6K signaling pathway
The distinctive feature of RA are synovial in ammation and synovial cell hyperplasia [36]. RA-FLSs, the key components of the invasive synovium, play an important role in the initiation and perpetuation of destructive joint in ammation [37]. The important chemokines (CXCL1, CXCL5,, G-CSF, etc.), in ammatory cytokines (TNF-a, IL-1b, IL-6, etc.) and in ammatory mediators (TLR-2, TLR-4, iNOS, etc.) released by RA-FLSs can promote the in ltration of DCs, monocytes, neutrophils, macrophages, B cells and T cells into joints, which lead to the chronic in ammation and joint destruction [36]. Recent studies have demonstrated that the mammalian target of rapamycin/p70 ribosomal protein S6 kinase (mTOR/p70S6K) signaling pathway has a major role in regulating cell survival and apoptosis, which is over-activated in RA-FLSs, as in Fig. 5a [38]. mTOR is upregulated in a variety of cancers and regulates cancer cell invasion, and its expression also related to the poor prognosis in cancer [24,37,[39][40][41][42]. It is well known that the 70 k Da ribosomal S6 kinase p70S6K1 regulates cell growth by inducing protein synthesis components [40]. Considering this background information, we hypothesized that the apoptosis-inducing effect of hybrid hydrogels on RA-FLSs might be correlated with the inhibition of the mTOR/p70S6K signaling pathway. Therefore, we analyzed the protein expressions levels of total mTOR p70S6K p-mTOR p-p70S6K in the each group by western blotting analysis (Fig. 5b and Fig. 5c). The results showed that hydrogels-treated with RA-FLSs, which were exposed to 10 min NIR irradiation at 0.38 W/cm 2 after administered, the levels of phosphorylated mTOR had signi cantly reduced (Fig. 5b, phospho-mTOR/total median reduction of 54%), as well as reduced the level of a phosphorylated form of one mTOR substrate, p70S6K (Fig. 5c, phospho-p70S6K/total median reduction of 38%), con rms that the inhibition of this pathway is correlated with reduction of apoptosis. These data indicated that hybrid hydrogels can inhibit the mTOR/p70S6K signaling pathway in RA-FLSs.

In vivo targeted effects and NIR imaging
CIA mice were generated to evaluate the targeted effects and NIR imaging of TP-Au/HA hybrid hydrogels on in ammatory joints in vivo. These mice were intraarticular injected with TP-Au/HA hybrid hydrogels solution (200µl). Because of superior capabilities of Au nanoshell, the hybrid hydrogels will lead to composite systems of hydrogels with unique optical properties [43,44]. The injected hybrid hydrogels were monitored in vivo NIR absorbance at 5min and 24h (Fig. 4a). We noted that the color of the in ammatory paws was changing, while observed the color of nonin ammatory paws was not obvious changing, which indicated that nanoparticles were selectively delivered to the in ammatory areas and accumulated in the in ammatory joints. Meanwhile, we noted the color of the in amed paw of CIA mice changing over time because of the localization of TP-Au-RGD nanoparticles in the in ammatory paws (Fig. 4b). These results indicated that TP-Au-RGD nanoparticles were preferentially delivered to the in ammation region by active targeting, rather than passive targeting, and accumulated in the in ammatory paws effectively.

In vivo e cacy
In order to study the treatment e cacy of hybrid hydrogel, divide CIA mice into four groups (n = 5 mice each group) with intraarticular administration, as summarized in Table 1. Figure 4e shows the clinical index of these groups over time. 35 mg/kg TP was injected intraarticularly four times a week (G2) as a positive control. Compared to the mice treated with saline (G1), other groups showed the decrease of clinical indices observed with some variations depending on the day. In G3, the decrease of clinical indices was slow until about day 20 and then increased, which the mice treated with TP-Au/HA hybrid hydrogels (0.2 mg/kg) with non-NIR irradiation, though they were lower than G1. However, may be due to photothermally controlled drug release, the clinical indices of G4 were lower than G2, which the administration of CIA was the TP-Au/HA hybrid hydrogels (0.2 mg/kg) combined with 10 min 1.59 W/cm 2 NIR irradiation. As shown in Fig. 4d, the release of TP for G3 is slow; therefore, the lack of NIR light impeded the sustained release of TP, resulting in the poor irradiation effects. In contrast, when the in ammatory paws exposed to NIR light (G4), the release of more than 10% of the TP within 12 h (Fig.  4d), which is a much lower TP dosage than G2 and obtained high therapeutic effect. These results indicate that photothermal-chemo therapy using TP-Au/HA hybrid hydrogels is a good way in the treatment of RA, which can maximize irradiation e cacy and reduce dose-related TP side effects. Administrated content was injected intraarticularly by volume 200µL. The joints was exposed to NIR light for 10 min with 1.59 W/cm 2 after injection.

Histopathological evaluation
To con rm the effects of the targeted photothermal-chemo treatment, joints histological examinations were performed 28 days after intraarticular injection (Fig. 6a). Severe in ltration of in ammatory cell was showed in the joint sections of untreated mice. The in ltration of in ammatory cell was signi cantly reduced in mice injected with TP solution four times a week (G2) and TP-Au/HA hybrid hydrogels with NIR irradiation (G4). In contrast, signi cant differences were not observed in G3 mice (Fig. 6b).

CT imaging
We performed three-dimensional micro-CT to assess bony changes in the paws of CIA mice (Fig. 6c). The severe bone destruction of paws, which treated CIA mice with saline. The bone structures of G2 and G4 were relatively well preserved. The bone volume of the paws of CIA mice was measured to determine the extent of bone preservation. The bone volume of the paws tended to be better preserved, which treated with TP-Au/HA hybrid hydrogels and NIR irradiation (Fig. 6d). These results indicated that the therapeutic effect of TP-Au/HA hybrid hydrogels with NIR irradiation is comparable to conventional TP treatment.
3.9 In vivo toxicity of TP-Au/HA hybrid hydrogels Histological examinations of major organs (liver, spleen, lung, kidney and heart) were performed 28 days after injection to investigate the in uence of TP-Au/HA hybrid hydrogels on the major organs (Fig. 4f). G4 showed no apparent tissue damage compared to the healthy control group, demonstrating that the TP-Au-RGD nanoparticles did not accumulated in major organs and induce toxicity in vivo.

Conclusions
In summary, we demonstrate the synergistic therapeutic effects of TP-Au/HA hybrid hydrogels with chondroprotection, anti-in ammation, photothermal-chemo therapy and in vivo imaging for the treatment of RA. When the hybrid hydrogels were administrated into the joints, the in vivo NIR absorbance images showed that the nanoparticles can selective accumulate in the in ammation areas of CIA mice. The release rate of TP from the hybrid hydrogels has accelerated because of NIR irradiation, allowing the photothermal-chemo therapy. Compared to conventional single treatment with TP, the treatment of TP-Au/HA hybrid hydrogels combined with NIR irradiation had excellent treatment effects with a smaller dosage of TP and low toxicity. Furthermore, the ndings suggested that TP-Au/HA hybrid hydrogels inhibited in ammation via suppressing the invasion and migration of RA-FLSs, by blocking the phosphorylation of the mTOR/p70S6K pathway at least in part. The result demonstrates that the targeted photothermal-chemo therapy using hybrid hydrogels in the treatment of RA is an effective strategy that can maximize the therapeutic effects and reduce dose-related side effects.
Declarations ACKONWLEDGMENTS Not applicable.  The preparation route of TP-Au/HA hybrid hydrogels.      (b) TP-Au/HA hybrid hydrogels reduced levels of phospho-mTOR, with a phospho-mTOR/total median reduction of 54% and (c) Levels of phosphorylation of mTOR targets phospho-p70S6K (phospho-p70S6K/total median reduction of 38%, con rms that the inhibition of this pathway. (d) Effect of TP-Au/HA hybrid hydrogels on the anti-proliferation of RA-FLSs. The data are expressed as the mean ± SEM values. By comparison with blank group,*P 0.05.