Materials

We used the following materials: CCK-8 assay kit (Beyotime, C0037), ELISA kits for IL-1β, IL-6, and TNF-α (Proteintech, KE10003, KE10091, KE10002), Strand cDNA Synthesis Super Mix kit (Yeasen, 11119ES60), qPCR SYBR Green Master Mix (Yeasen, 11203ES08), Tacrolimus (Aladdin, T101160), DAPI (Beyotime, C1006), SF488-Phalloidin (Solarbio, CA1640), and Cy5.5 (TOPSCIENCE, TD0091). Supplementary Tables 1 and 2 provide further details on antibodies and primer sequences.

Cells and animals

We obtained RAW 264.7, MH7A, and HUVEC cells from the American Type Culture Collection (ATCC). These cells were cultured in Dulbecco’s minimum essential medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco), 1% penicillin (100 IU/mL, Corning), and streptomycin (100 µg/mL, Corning). Eight-week-old male DBA/1J mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were kept under a 12-hour light/dark cycle at a temperature of 25 ± 2 °C and humidity of 60 ± 10%. The Institutional Animal Care and Use Committee of the Chinese Academy of Sciences approved the animal experiment protocols, which complied with relevant ethical regulations.

Preparation and characterization of HA-NG/TAC

A mixture of 1-hexylamine and N-carboxy anhydrides of L-lysine (NCA) at a molar ratio of 1:6 was combined in a flame-dried flask and dissolved in anhydrous DMF. The blend was agitated at 25 °C for 72 h, resulting in ring-opening polymerization (ROP) of L-lysine NCA, facilitated by 1-hexylamine as the macroinitiator. L-lysine and L-cystine were then added at a molar ratio of 10:5, and the mixture was agitated at 25 °C for another 72 h to obtain NG (1-hexamine-poly(L-Lysine-co-L-Cystine)). The nanogel precipitate was washed twice with diethyl ether, dried under vacuum for 72 h at room temperature, dissolved in trifluoroacetic acid (TFA), and mixed with hydrobromic acid/acetic acid (HBr/Hac) (33 wt%). After stirring for 1.5 h at 25 °C, the mixture was dialyzed in a dialysis bag (MWCO 500 Da) for 24 h and freeze-dried. NG/TAC was obtained by mixing 1-hexanamine-poly(L-Lysine-co-L-Cystine) with TAC in DMF and agitating for 24 h. HA-NG/TAC was produced by mixing NG/TAC with HA until the solution became homogeneous and transparent. The morphology of HA-NG/TAC was observed using transmission electron microscopy (TEM). Dynamic laser scattering (DLS) measured the hydrodynamic sizes and zeta potentials of NG/TAC and HA-NG/TAC.

In vitro TAC loading and release from HA-NG

We calculated the drug loading content (DLC%) of TAC-loaded HA-NG using the equation:

$$\:DLC\%\:=\:\frac\:\times\:100\%$$

The release behaviors of HA-NG/TAC were investigated in vitro using PBS with 0.5% (v/v) Tween 80 solution at pH 7.4. Freeze-dried HA-NG/TAC was suspended in 5 mL of PBS with 0 or 10 mM GSH inside a dialysis bag (MWCO 3500 Da). During the release trial, the dialysis bag was submerged in 50 mL of PBS with either 0 or 10 mM GSH and agitated at 70 rpm. At predetermined intervals, 2 mL of the external release medium were extracted and replaced with an equal volume of fresh medium. The released drug amount was measured by high-performance liquid chromatography (HPLC) using standard curves at 311 nm.

Actively targeting of HA-NG/TAC in vitro

To establish an inflammatory condition, RAW 264.7 cells were stimulated with lipopolysaccharide (LPS) for 24 h, while untreated macrophages served as the control group. Subsequently, macrophages were incubated with Cy5.5-NP and Cy5.5-labeled HA-NG for 4 h. Cells were fixed with 4% paraformaldehyde for 15 min, treated with SF488-Phalloidin for 15 min, followed by a 5-minute incubation with DAPI, and then analyzed using confocal laser scanning microscopy (CLSM). The fluorescence intensity was measured using ImageJ software. RAW 264.7, MH7A, and HUVEC cells were cultured for 24 h and stimulated with LPS for an additional 24 h. The cells were then incubated with Cy5.5-labeled HA-NG for 4 h, fixed for 15 min, stained with DAPI, and observed by CLSM. Fluorescence intensity was analyzed using ImageJ software.

Toxicity of HA-NG/TAC in vitro

RAW264.7 cells were cultured at 37 °C for 24 h. Different concentrations of NG/TAC, HA-NG/TAC, and TAC were administered and incubated for 24 h. Additionally, RAW 264.7, MH7A, and HUVEC cells were cultured for 24 h. HA-NG/TAC was introduced and incubated for 48 h and 72 h, respectively. After adding the CCK-8 reagent and incubating at 37 °C for 2 h, the optical density at 450 nm was measured using a microplate reader.

Hemolysis test

Fresh blood was centrifuged, and the serum was removed. The blood cells were dispersed in 4 mL of NaCl solution and centrifuged at 1500 rpm for 10 min. Next, 200 µL of this solution was diluted with NaCl to 10 mL. HA-NG/TAC was added to the blood cells to achieve final concentrations ranging from 0.01 µg/mL to 100 µg/mL through serial dilutions. Additionally, washed blood was dissolved in 1 mL of water as a positive control and in 1 mL of NaCl as a negative control. All solutions were incubated in a 37 °C water bath for 2 h, then centrifuged at 6000 rpm for 10 min. The states were photographed and recorded, and the supernatant was collected into a quartz cuvette. Absorbance at 540 nm was measured using a UV spectrophotometer, and the hemolysis rate was calculated using the formula:

$$\:Hemolysis=\:\frac_-_}_-\:_}\:\times\:100\%$$

where Abspositive control is the absorbance of the positive control group, Absnegative control is the absorbance of the negative control group, and Abssample is the absorbance of the test sample group.

Western blot

CD44 was extracted from RAW 264.7, MH7A, and HUVEC cells using RIPA buffer with 1 µM PMSF. Protein concentrations in the supernatant were measured using a BCA protein assay kit and analyzed by Western blot. After separation by 10% SDS-PAGE, the proteins were transferred onto PVDF membranes. The membranes were blocked with Blocking Buffer for Western Blot at room temperature for 10 min and incubated with primary antibodies for CD44 (Proteintech, 15675-1-AP) at 4 °C overnight. They were then treated with HRP-labeled secondary antibodies at room temperature for 1 h. Finally, the bands were detected using an ECL reagent and visualized with a chemiluminescence imaging system.

2.9. Establishment of collagen-induced arthritis mouse model

The CIA mouse model was established according to a previous study using eight-week-old male DBA/1J mice [27]. Bovine type II collagen was emulsified with complete Freund’s adjuvant containing 2 mg/mL of Mycobacterium tuberculosis. The mixture was subcutaneously administered to DBA/1 mice at the base of the tail. After 21 days from the initial injection, the mice received an additional injection of type II collagen mixed with incomplete Freund’s adjuvant near the primary injection site.

Specific distribution of HA-NG/TAC in vivo

After reaching an average clinical score of 14, the mice with CIA were injected with Cy5.5-labeled NG and Cy5.5-labeled HA-NG via the tail vein. The in vivo imaging system was used to observe the fluorescence image in vivo at the 0.5th, 3th, 6th, 12th, 24th hour post-injection. Subsequently, the mice were euthanized, and their vital organs, such as the heart, liver, spleen, lungs, kidneys, and joints, were collected and examined using the in vivo imaging system. The inflamed paws were fixed in 4% paraformaldehyde and decalcified using a 20% (w/v) ethylenediaminetetraacetic acid (EDTA) solution, with daily changes for 30 days. Subsequently, the joints were sectioned, stained with AF488-CD86 antibody and DAPI, and examined under fluorescence microscopy. ImageJ software was utilized for quantitative analysis of the fluorescence intensity.

Therapeutic efficacy of HA-NG/TAC in vivo

Following the second immunization, CIA mice were established and evaluated every other day in accordance with the aforementioned protocol. We assessed every mouse individually using a scale of 0 to 4. 0, no evidence of erythema and swelling; 1, erythema and mild swelling confined to the tarsals or ankle joint; 2, erythema and mild swelling extending from the ankle to the tarsals; 3, erythema and moderate swelling extending from the ankle to metatarsal joints; 4, erythema and severe swelling encompass the ankle, foot and digits, or ankylosis of the limb. Every mouse was given a total arthritis score based on its four paws. Once the model was established, the mice were divided into four groups at random and given PBS, TAC, NG/TAC, and HA-NG/TAC for treatment according to their clinical scores. Healthy DBA mice were treated as a control group. In CIA mice, intravenous injections were given every two days after the 30th day of primary immunization. Following a period of 48 days after the primary immunization, the mice were sacrificed.

Histological and immunofluorescent assay

Following the sacrifice of the arthritis mice, the organs (heart, liver, spleen, lung, kidney) and hind knee joints were collected and combined in a solution of 4% paraformaldehyde. Subsequently, the joints were decalcified for a period of 30 days using a 20% (w/v) ethylenediaminetetraacetic acid (EDTA) solution, and then sliced. Subsequently, the tissue slides underwent staining with H&E. Two researchers evaluated the Histological Synovitis Scores (HSS) (Supplementary Table 3), and the outcomes were determined based on the mean scores. The tissue sample slides were blocked with 5% BSA and treated with antibodies targeting IL-1β, IL-6, and TNF-α. To identify these pro-inflammatory cytokines in the tissue slides, a fluorescence microscope was utilized.

Micro-CT reconstruction and analysis

We used Mimics software to generate 3D images of the hind paw joints by scanning them with a Micro-CT scanner at a resolution of 15 μm. A quantitative analysis of bone mineral density (BMD), the rate of bone volume and total volume (BV/TV), trabecula number (Tb. N), trabecula spacing (Tb. Sp), and trabecula thickness (Tb. Th) was performed using CTAn software.

Expression level of pro-inflammatory cytokines in vivo

ELISA kits were used to detect the serum levels of IL-1β, IL-6, and TNF-α in each group, following the manufacturer’s instructions. To evaluate the samples, a microplate reader was utilized with wavelengths of 450 nm and 630 nm. After crushing the hind knee joints of CIA mice in liquid nitrogen and extracting the supernatant, we obtained total RNA using TRIZOL. Afterwards, the trichloromethane and isopropanol were used to purify the total RNA. Strand cDNA Synthesis Super Mix kit instructions were followed to synthesize cDNA. IL-1β, IL-6, and TNF-α mRNA were successfully detected through a qPCR experiment utilizing qPCR SYBR Green Master Mix and an Agilent MX3000P instrument. GAPDH was used as a reference for standardized mRNA expression.

Statistical analysis

Each experiment required a minimum of three trials. GraphPad Prism 9 was used to calculate the mean value ± standard error of mean (SEM) based on the given number of experiments. An unpaired Student’s t-test was used to compare two groups, while a one-way analysis of variance was used to analyze more than two groups. Significant differences were considered when p-value was less than 0.05.