Preparation and characterization of CS-Cur-TGNCs

TEM is a means of observing the morphological structure of nanoparticles. Figure 1A and B show that CS-Cur-TGNCs were regular monodisperse solid spheres approximately 80 nm in size, with a distinct core-shell structure and an inner core approximately 75 nm in size and an outer capsule wall material thickness of approximately 5 nm. DLS revealed that the sizes of the Cur-NCs, Cur-TGNCs and CS-Cur-TGNCs were 42.29 ± 5.42 nm (PDI: 0.232 ± 0.071), 80.05 ± 18.7 nm (PDI: 0.301 ± 0.054) and 80.80 ± 16.47 nm (PDI: 0.337 ± 0.049), respectively, with zeta potentials of -12.66 ± 3.52 mv, -24.70 ± 2.61 mv and − 35.87 ± 5.90 mv, respectively (Fig. 1C and D and S1). Compared with Cur-TGNCs, CS-Cur-TGNCs exhibited a slight increase in particle size and a significant decrease in potential, which was related to the strong negative charge of CS. FTIR spectroscopy was used to further verify whether CS was successfully grafted on the surface of Cur-TGNCs, and the results were shown in Figure S2. The stretching vibration of the lipid sulfate group (S = O) was present at 1020 cm− 1 for CS, while the characteristic peak of CS was absent for Cur-TGNCs. Compared with Cur-TGNCs, CS-Cur-TGNCs showed the lipid sulfate group and C-O stretching vibration of CS at 1205 cm− 1 and 1029 cm− 1, which indicated that CS had been successfully modified on the surface of Cur-TGNCs. Subsequently, we observed the stability of CS-Cur-TGNCs in different physiological solutions (water, PBS, 0.9% NaCl, DMEM, and 10% FBS). The results in Fig. 1E showed that CS-Cur-TGNCs had good dispersion in all media. After 7 d, CS-Cur-TGNCs exhibited the phenomenon of sedimentation in all solutions, and it was able to disperse uniformly after uniform shaking, indicating its excellent dispersibility. The changes in particle size, potential and PDI values were detected at 0 d and 7 d, and the results showed that the CS-Cur-TGNCs in different media did not change significantly within seven days and exhibited good stability (Fig. S3). The stability of Cur was poor, so the change in the stability of Cur in the CS-Cur-TGNCs was assessed via drug stability tests. The degradation rates of CS-Cur-TGNCs and free Cur in PBS (37 °C, pH = 7.4) after 6 h were approximately 15% and 40%, respectively, and the degradation rate of Cur in the CS-Cur-TGNCs was 0.38 times greater than that in the free Cur (Fig. 1F). The CS-Cur-TGNCs significantly improved the stability of Cur. The crystallinity and thermal properties of Cur, Cur-TGNCs and PM were observed by PXRD and DSC. The PXRD curves showed that Cur is a crystalline drug with multiple diffraction peaks, and strong crystal diffraction peaks appeared in the range of 3° to 50°. The positions of the characteristic diffraction peaks of Cur, PM and Cur-TGNCs did not change significantly, indicating that Cur existed in a nanocrystalline form in Cur-TGNCs. Compared with that of Cur, the intensity of the diffraction peak of Cur-TGNCs was significantly lower (Fig. 1G). The crystallinities of Cur and Cur-TGNCs were calculated to be 52.82% and 36.59%, respectively, using Origin software (originpro 2023). The crystallinity of Cur was significantly lower in Cur-TGNCs than in other materials, and drugs with low crystallinity were more favorable for dissolution and absorption. This result was further verified by the DSC results (Fig. 1H). The characteristic absorption peaks of the Cur, PM and Cur-TGNC samples appeared near 176 °C, and compared to those of Cur, the intensities of the absorption peaks of Cur-TGNCs were significantly lower. This is related to the fact that Cur exists in the form of nanocrystals in Cur-TGNCs. According to the Ostwald freundlich equation and Noyes-Whitney equation, the smaller particle size and larger specific surface area of Cur nanocrystals can increase the solubility and dissolution rate of Cur, thereby improving the bioavailability and efficacy of the Cur. Next, we determined the in vitro release of Cur from the CS-Cur-TGNCs. As shown in Fig. 1I, in the absence of MMP-2, 40% of the CS-Cur-TGNCs were released after 12 h, indicating that the release of these compounds was significantly slowed. In the presence of MMP-2, 90% of the CS-Cur-TGNCs were released after 12 h. This finding suggested that CS-Cur-TGNCs are able to rapidly release Cur in the inflamed joints of RA patients (in which MMPs are highly expressed), which in turn increases drug bioavailability and improves therapeutic efficacy. This finding is supported by the results of pharmacokinetic experiments; the t1/2 values of Cur and CS-Cur-TGNCs were 0.06 ± 0.01 h and 0.40 ± 0.15 h, respectively. The AUC0-t values of Cur and CS-Cur-TGNCs were 56.24 ± 12.97 ng/L*h and 344.49 ± 147.37 ng/L*h, respectively. Compared with Cur, CS-Cur-TGNCs had approximately 7-fold and 6-fold greater t1/2 and AUC0-t values, respectively (Fig. 1J). These findings indicated that CS-Cur-TGNCs significantly improved the bioavailability of Cur. Hemolysis and cytotoxicity assays were used to study the biosafety of the CS-Cur-TGNCs. The results of the hemolysis assay showed that the CS-Cur-TGNCs did not exhibit hemolysis in the range of 125–1000 µg/mL (Fig. 1K, L). CCK-8 assays revealed that the viability of RAW264.7 cells cultured in the presence of 80 µg/mL CS-Cur-TGNC was greater than 80% (Fig. 1M). The results of hemolysis and cytotoxicity assays indicated that the CS-Cur-TGNCs had good biosafety. Based on these results, we believe that CS-Cur-TGNCs have potential as a novel drug delivery system for the treatment of arthritis.

Fig. 1

Preparation and characterization of CS-Cur-TGNCs. (A) The TEM imaging of CS-Cur-TGNCs. (B) The magnified TEM imaging of CS-Cur-TGNCs. (C) The particle size of Cur-NCs, Cur-TGNCs and CS-Cur-TGNCs. (D) The zeta potential of Cur-NCs, Cur-TGNCs and CS-Cur-TGNCs. (E) Images of CS-Cur-TGNCs in different buffers at 0 d and 7 d (from left to right, water, PBS, 0.9% NaCl, DMEM, and 10% FBS) as well as resuspension after 7 d (right panel). (F) Cur Variation of concentration over time. (G) The PXRD patterns of Cur, Cur-TGNCs and PM (physical mixture of Cur, tragacanth gum and gelatin). (H) The DSC spectra of Cur, Cur-TGNCs and PM (physical mixture of Cur, tragacanth gum and gelatin). (I) In vivo drug release curves of Cur and CS-Cur-TGNCs. (J) The plasma concentration-time curve of Cur and CS-Cur-TGNCs. (K, I) The hemolysis experiment diagram and hemolysis rate of CS-Cur-TGNCs. (M) The cytotoxicity of CS-Cur-TGNCs on RAW264.7 cells

Cell uptake and targeting ability of the CS-Cur-TGNCs

CD44 expression at the junction sites was detected via Western blot experiments, and the ability of CS-Cur-TGNC to target cells and tissues was assessed via cellular uptake analysis and in vivo imaging systems. Western blot analysis (Fig. 2A, B) revealed a significant increase in CD44 expression in the joint area of CIA mice compared with that in the sham group (approximately 1.7-fold). CLSM was applied to determine the ability of the RAW264.7 cells to take up free Cur, Cur-TGNCs and CS-Cur-TGNCs. As shown in Fig. 2C, the intensity of uptake of Cur-TGNCs by RAW264.7 cells was significantly greater than that of Cur, and the fluorescence level was approximately 4.5 times greater than that of Cur (Fig. 2D). As shown in Fig. 2E, the fluorescence of CS-Cur-TGNCs was greater than that of Cur-TGNCs, indicating that RAW264.7 cells had a greater uptake effect on the CS-Cur-TGNCs. This is because the sulfated groups of CS can interact with the CD44 receptor on the surface of macrophages, and the negative charge of the sulfated groups and the amino glucose units can also stabilize the interaction between CS and CD44 [37, 38]. Moreover, Cur (green light) exhibited a high degree of consistency with Cur-TGNCs and CS-Cur-TGNCs (red light), suggesting that the increase in Cur absorption is closely related to the structural properties of the CS-Cur-TGNCs. The same phenomenon was observed using TEM (Fig. 2F). CS-Cur-TGNCs and Cur-TGNCs were taken up by activated RAW264.7 cells mainly by endocytosis. Due to the ability of CS to target CD44, RAW264.7 cells exhibited increased uptake of CS-Cur-TGNCs. More CS-Cur-TGNC nanoparticles were observed inside the cells. We next used lysosomal staining to assess the intracellular localization of the CS-Cur-TGNCs in RAW264.7 cells. CS-Cur-TGNCs (green) colocalized well with LysoTracker (red) (Fig. 2G). The colocalization coefficients at 2 h and 4 h were 0.41 and 0.52, respectively. Therefore, we inferred that the lysosomal structure within RAW264.7 cells mediates the intracellular internalization of CS-Cur-TGNCs, which is a common phenomenon in the cellular uptake of nanoparticles. The effect of energy on the uptake of the CS-Cur-TGNCs by RAW264.7 cells was explored by performing cell uptake experiments at 4 °C and 37 °C. The results showed that the intensity of CS-Cur-TGNC uptake by RAW264.7 cells was significantly greater at 37 °C than at 4 °C (Fig. 2H). These findings indicate that the RAW264.7 cell-mediated cellular internalization of CS-Cur-TGNCs was energy dependent. To evaluate the joint targeting ability of the CS-Cur-TGNCs, as shown in Fig. 2I, we established a CIA mouse model using the repeat immunization method. At 35 days, joint images of the mice at different time points were obtained using an in vivo imaging system (IVIS). The results showed that Cur-TGNCs were able to accumulate at the joint site. Compared with that of Cur-TGNCs, the fluorescence intensity of CS-Cur-TGNCs was significantly greater, with the highest fluorescence occurring at 8 h (Fig. 2G and L). It has been well demonstrated that the use of CS-Cur-TGNCs as a nanodelivery system for targeting joints can achieve accumulation at the joint site, which will improve the therapeutic efficacy of Cur. In addition, among the important metabolic organs (heart, liver, spleen, lungs, and kidneys) in mice, the liver and kidneys exhibited increased fluorescence intensity (Fig. 2K and M). These results indicated that the CS-Cur-TGNCs were metabolized mainly by the liver. This is similar to the metabolic phenomenon of most nanoparticles [39]. The liver as main metabolic organ can metabolism and clear the foreign substances via the process of internalization by hepatic cells such as Kupffer cells and liver endothelial cells. These results suggest that CS-Cur-TGNCs can be taken up by macrophages and accumulate at the joint site.

Fig. 2

Cell uptake and targeting validation of CS-Cur-TGNCs. (A, B) Expression and quantification of CD44 in the joints of healthy mice and CIA mice. (C, D) The uptake of Cur and Cur-TGNCs by RAW264.7 cells after inflammatory activation and quantitative statistics. (E) The uptake of Cur-TGNC and CS-Cur-TGNCs by RAW264.7 cells after inflammatory activation and quantitative statistics. (F) Cell morphology and mitochondrial morphology of each group under TEM; Sclar bar = 2 μm, 1 μm and 500 nm. (G) RAW264.7 on the uptake of CS-Cur-TGNCs at various points in time. Red light (lysosomal probe). (H) RAW264.7 uptake of CS-Cur-TGNCs at different temperatures at 2 h. (I) Experimental outline. (J, L) Pictures and statistical analysis of the mouse IVIS system. (K, M) Pictorial and statistical analysis of the vital metabolic organs of the mouse IVIS system. Data are shown as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001

Exploring the mechanism of action of Cur in rheumatoid arthritis

To clarify the disease targets of Cur, we performed target prediction with the Swiss Target Prediction Database, and the prediction results were collated to obtain 64 drug targets. For RA disease-related targets, the DisGeNET and GeneCards databases were used to collate the predicted relevant targets with probable values greater than 2.5, from which 3340 RA disease-related targets were obtained. Cross-analysis of disease targets and predicted drug targets yielded 36 intersecting targets (Fig. 3A and B). To visualize the interactions between Cur and its targets, cross-targets and Cur were imported into Cytoscape 3.8.2 software to construct a component-disease target interaction network (Fig. 3C). To analyze the target interactions, 36 intersecting targets were imported into STRING, and “Homo sapiens” was selected as the “Species” to construct a PPI network (Fig. 3D), which consisted of a total of 36 nodes (representing the interacting targets) and 138 edges (representing the interaction relationships between the targets). As shown in the figure, the core targets we obtained were mainly ALPLCA1, MMP8, ABCC1, ADAM17, AGTR1, AKT1, APPAURKA, BCL2, BRAF, CSF1R, CXCR2, EGFR and other genes, which implies that these targets play important roles in the network. GO functional annotation of the core targets of curcumin in RA was performed, and the results are shown in Fig. 3E. Larger bubbles represent more genes enriched in that GO entry, and redder bubbles indicate more significant GO features. The obtained GO-enriched terms were enriched mainly in the terms “positive regulation of peptidyl-serine phosphorylation”, “positive regulation of cell growth”, “positive regulation of NIK/NF-κB signaling”, “inflammatory response” and “response to oxidative stress”. In this study, we used network pharmacology to explore the molecular mechanism of Cur in the treatment of RA and to provide a reference for subsequent studies.

Fig. 3

Exploring the mechanism of action of Cur in rheumatoid arthritis. (A) and (B) Venn diagrams and crossover plots of active ingredients for Cur and RA disease targets. (C) Network plot of Cur-RA target interactions. (D) PPI network plots of crossover targets in the STRING database. (E) Target GO/KEGG enrichment analysis plot

Antioxidant stress activity of the CS-Cur-TGNCs

The antioxidant effect of CS-Cur-TGNCs was detected using an oxidative stress-related kit (ABTS kit, KI reagent, MDA kit, SOD kit and GSH kit) and cellular immunofluorescence. The ABTS assay is often used to detect the in vitro antioxidant capacity of substances. When ABTS is oxidized to a stable blue‒green ABTS + radical, there is a characteristic absorption peak at 405 nm. The presence of an antioxidant inhibits the production of ABTS+, resulting in a lighter color and lower absorbance of the solution. Figure 4A shows that at concentrations of CS-Cur-TGNCs ranging from 0 to 200 µg/mL, the color of the solution gradually changed from blue‒green to light blue‒green, and the absorbance values also changed significantly. These results indicated that the ROS scavenging ability of the CS-Cur-TGNCs was concentration-dependent. As the concentration of CS-Cur-TGNCs increased, the absorption intensity of I2/I3− at 350 nm gradually decreased. The lowest absorption density was observed at a CS-Cur-TGNCs concentration of 200 µg/mL, which further indicated that the scavenging effect of CS-Cur-TGNCs on H2O2 was significantly concentration dependent (Fig. 4B). Next, we examined the antioxidant effects of CS-Cur-TGNCs at the cellular level using MDA, SOD, and GSH assays. Among the groups, the RAW264.7 cells had the highest level of MDA after inflammatory activation and the lowest level of MDA in the CS-Cur-TGNCs group; these values were approximately 0.56-fold greater than those in the LPS group (Fig. 4C). In addition, the levels of SOD and GSH were the lowest after LPS stimulation, and the levels gradually increased in the different treatment groups, with the highest levels occurring in the CS-Cur-TGNCs group (Fig. 4D, E). These results again demonstrated that CS-Cur-TGNCs have excellent antioxidative stress effects. Cellular fluorescence experiments using ROS fluorescent probes were used to evaluate the antioxidant activity of the CS-Cur-TGNCs. The results showed that the highest fluorescence intensity was observed in the LPS-induced group, and the fluorescence intensity gradually decreased in the Cur, Cur-TGNCs and CS-Cur-TGNCs groups. Among the groups, the CS-Cur-TGNCs group had the weakest fluorescence signal, indicating that the CS-Cur-TGNCs had a strong ability to reduce ROS levels (Fig. 4F, G). HIF-1α is a key indicator of oxidative stress, and changes in HIF-1α expression reflect the level of oxidative stress. We assessed the expression of HIF-1α in cells by cytofluorimetric assay. The results showed that the upregulated HIF-1α in RAW264.7 cells was reduced by the Cur, Cur-TGNCs and CS-Cur-TGNCs treatments, especially in the case of the CS-Cur-TGNCs treatment, which had a stronger effect (Fig. 4H, I). These results suggest that CS-Cur-TGNCs can alleviate oxidative stress, ameliorate hypoxia at joint sites, and play a therapeutic role in the treatment of arthritic diseases.

Fig. 4

Antioxidant stress activity of CS-Cur-TGNCs. (A) ROS scavenging ability of CS-Cur-TGNCs at different concentrations after reacting with ABTS solution. (B) The UV-Vis spectra of I2/I3- after KI were incubated with different concentrations of CS-Cur-TGNCs for 30 min. (C–E) MDA and SOD concentrations and GSH activity in different groups of RAW264.7 cells. (F, G) DCFH-DA probe fluorescence Images and quantitative analysis of different groups of RAW264.7 cells. (H, I) HIF-α fluorescence images and quantitative analysis of different groups of RAW264.7 cells. Data are shown as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001

Anti-inflammatory activity of the CS-Cur-TGNCs

To investigate the anti-inflammatory effects of CS-Cur-TGNCs, we evaluated the expression of an inflammatory factor (IL-1β) and the conversion of M1-type macrophages to M2 macrophages after different treatment modalities by immunofluorescence and q-PCR assays. The expression of IL-1β is a key inflammatory factor in arthritic diseases. The cell fluorescence results showed that the expression of IL-1β (green) was significantly increased in LPS-induced RAW264.7 cells. In comparison, the fluorescence intensity gradually decreased after Cur, Cur-TGNCs and CS-Cur-TGNCs treatments (Fig. 5A). Specifically, the fluorescence intensity in the CS-Cur-TGNCs group was approximately 0.63 times greater than that in the Cur-TGNCs group (Fig. 5D), suggesting that the CS-Cur-TGNCs play a significant role in suppressing inflammation. In addition, we examined the expression levels of the M1-type macrophage marker (iNOS) and M2-type macrophage marker by a cytofluorescence assay. The results showed that the M1-type macrophage marker (iNOS) was upregulated by LPS to different degrees after Cur and Cur-TGNCs treatments, and the most significant decrease was observed in the CS-Cur-TGNCs group (Fig. 5B, E). In contrast, the expression of the M2-type macrophage marker Arg-1 was significantly upregulated after CS-Cur-TGNCs treatment and was approximately 2.67-fold and 1.45-fold greater than that in the Cur and Cur-TGNCs groups, respectively (Fig. 5C, F). These results suggest that CS-Cur-TGNCs can promote the transition of M1 macrophages to M2 macrophages. Next, we verified the anti-inflammatory effects of CS-Cur-TGNCs at the animal level by tissue fluorescence, ELISA and q-PCR experiments. As shown in Fig. 5H, both Cur and Cur-TGNCs downregulated the expression of iNOS in the joint area, whereas the CS-Cur-TGNCs-treated group had the lowest fluorescence intensity, which was approximately 0.55-fold and 0.63-fold greater than that of the Cur and Cur-TGNCs groups, respectively (Fig. 5G). Treatment with CS-Cur-TGNCs significantly downregulated the expression of a proinflammatory factor (IL-6 and TNF-α) and upregulated the expression of an anti-inflammatory factor (IL-10) in serum (Fig. 5I-K). q-PCR was used to detect the expression of key inflammatory proteins. Consistent with the tissue fluorescence results, the mRNA expression levels of relevant inflammatory factors (IL-6, iNOS, TNF-α, and IL-1β) were significantly increased in the model mice. Compared with Cur and Cur-TGNCs, CS-Cur-TGNCs inhibited the mRNA expression of inflammatory factors more significantly (Fig. 5L-O). The above experiments demonstrated that CS-Cur-TGNCs have excellent anti-inflammatory effects in the treatment of RA.

Fig. 5

Anti-inflammatory activity of CS-Cur-TGNCs. (A, D) IL-1β fluorescence images and quantitative analysis of different groups of RAW264.7 cells. (B, E) iNOS fluorescence images and quantitative analysis of different groups of RAW264.7 cells. (C, F) Arg-1 fluorescence images and quantitative analysis of different groups of RAW264.7 cells. (G, H) iNOS fluorescence images and quantitative analysis of different groups of knee joints. (I-K) Inflammatory factors (TNF-α, IL-10 and IL-6) were detected in the serum of different groups of mice (L-O) The expression of mRNAs (iNOS, TNF-α, IL-1β and IL-6) in ankle tissues was detected by q-PCR. Data are shown as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001

In vivo therapeutic effects on RA

We followed the treatment protocol shown in Fig. 6A and recorded joint recovery (photographs, ankle thickness, clinical scores, body weight, exercise recovery, and pathological changes) to explore the therapeutic effects of CS-Cur-TGNCs in CIA mice. Photographs of the joints of CIA mice show marked redness, swelling and articular deformity, with complete swelling of the lateral ankle joints (Fig. 6B). Whereas Cur was able to slightly ameliorate these symptoms, CS-Cur-TGNCs significantly reduced these joint swelling phenomena, showing stronger therapeutic effects than Cur and Cur-TGNCs. Bone erosion and destruction are important pathological features of rheumatoid arthritis for assessing its severity. We performed H&E and toluidine blue staining for histologic analysis of knee joint tissues from different treatment groups. The results showed that the joint tissues of the normal group did not exhibit joint inflammation or cartilage damage, and there was a clear interface between the bone and cartilage tissues. The PBS-treated CIA mice exhibited severe cartilage tissue destruction and severe inflammatory infiltration, but the joint structure was not obvious. Joint inflammation and cartilage erosion were significantly reduced in the Cur and Cur-TGNC groups. Moreover, the efficacy of the CS-Cur-TGNCs treatment was significantly greater than that of the Cur and Cur-TGNCs treatments, but the difference was not significant (Fig. 6C). In addition, the results of toluidine blue staining also showed that more cartilage and proteoglycans were distributed on the joint surface in the CS-Cur-TGNCs-treated group than in the control group (Fig. 6D). These results indicated that CS-Cur-TGNCs had a satisfactory effect on RA treatment. Subsequently, we measured the hind paw thickness, inflammation score and body weight of the mice. Changes in hind paw thickness and inflammation score are important indicators of the therapeutic efficacy of RA. Compared with that in the PBS-treated model group, the hind paw thickness was reduced in the Cur, Cur-TGNCs, and CS-Cur-TGNCs groups. Notably, the CS-Cur-TGNCs-treated group exhibited a more significant difference in cytokine levels (Fig. 6E). Moreover, CS-Cur-TGNCs had the same effect on reducing joint inflammation scores and significantly suppressing joint inflammation in CIA mice (Fig. 6F). Body weight is often used as an indirect indicator of RA recovery. After the injection of CS-Cur-TGNCs, the body weight of the CIA mice significantly increased and almost reached normal values (Fig. 6G). In addition, we evaluated the duration, distance, and movement speed of the mice that were stationary within 30 min in the different treatment groups. The CS-Cur-TGNCs treatment significantly improved the mobility deficits of the CIA mice (Fig. 6H-J). Therefore, we concluded that CS-Cur-TGNCs have a favorable effect on joint recovery.

Fig. 6

In vivo Therapeutic Effects on RA. (A) Establishment of the CIA mouse model and summary of experiments. (B) The representative photographs of the posterior ankle joints of different groups of CIA mice. (C) H&E images of knee joints from different groups of CIA mice. (D) Toluidine blue images of the knee joints of different groups of CIA mice. (E-G) Changes in hind paw thickness, treatment scores and body weight in different groups of CIA mice. (H-J) Time at rest, distance traveled, and speed of movement for 30 min in different groups of CIA mice. Data are shown as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001

In vivo therapeutic effects of GA

To further evaluate the therapeutic effects of CS-Cur-TGNCs on other arthritic diseases, we established a rat GA model according to the scheme shown in Fig. 7A. The joint targeting and inhibition of inflammation by CS-Cur-TGNCs were also evaluated via an in vivo fluorescence imaging system (IVIS), ELISA and q-PCR assays. Figure 7B shows that the fluorescence in the joint tissues of the Cur-TGNC treatment group was weaker than that in the control group, indicating that only a small amount of Cur-TGNCs accumulated at the site of inflammation. Compared with those of Cur-TGNCs, the fluorescence intensity of CS-Cur-TGNCs was greater at different time points (2, 4, 8, 12 and 24 h), and the fluorescence intensity reached a maximum at 8 h. Thereafter, the fluorescence intensity at the joint site gradually weakened with increasing time (Fig. 7E). At 8 h, the fluorescence levels in the liver and kidney of the CS-Cur-TGNCs- and Cur-TGNC-treated groups were significantly greater than those in the other organs (Fig. 7C, D). These results indicated that the CS-Cur-TGNCs and Cur-TGNCs were metabolized mainly by the liver and kidney, which was also consistent with the metabolic characteristics of the majority of the nanoparticles. Next, we assessed the levels of inflammatory factors (IL-6 and IL-10) in cell supernatants and rat serum, as well as the mRNA expression levels of inflammatory factors associated with the ankle joint site (IL-6, TNF-α, and IL-1β), using ELISA and q-PCR. Treatment with CS-Cur-TGNCs significantly downregulated the expression of a proinflammatory factor (IL-6) and upregulated the expression of an anti-inflammatory factor (IL-10) in cell supernatants (Fig. 7F, G). The results for the relevant inflammatory factors in the serum were in good agreement with the above results (Fig. 7H, I). Similar experimental results were obtained by q-PCR. The mRNA expression of proinflammatory factors (IL-6, TNF-α, and IL-1β) in the ankle joint was downregulated in the Cur, Cur-TGNCs, and CS-Cur-TGNCs treatment groups. CS-Cur-TGNCs downregulated the expression of these mRNAs to the most significant extent (Fig. 7J-L). These results suggest that, as a nanomedicine, CS-Cur-TGNCs can be applied not only for the treatment of RA but also for the treatment of GA, as they have excellent joint targeting ability and inflammation.

Fig. 7

In vivo therapeutic effects of GA. (A) Establishment of the GA rat model and summary of experiments. (B, E) Pictures and statistical analysis of the rat ankle IVIS system. (C, D) Pictures and statistical analysis of the vital metabolic organs of the rat ankle IVIS system. (F-I) Inflammatory factors (IL-6, IL-10) were detected in cell supernatants and rat serum of different groups. (J-L) The expression of mRNAs (TNF-α, IL-1β, and IL-6) in the ankle tissues of GA rats was detected by q-PCR. Data are shown as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001

CS-Cur-TGNCs promotes joint recovery in GA

To further investigate the efficacy of CS-Cur-TGNCs on GA, the joint recovery of GA rats was statistically analyzed. We observed that CS-Cur-TGNCs minimized paw swelling and inflammation (Fig. 8A). In addition, inflammation led to an increase in local temperature, and treatment with CS-Cur-TGNCs significantly reduced the local temperature to within the normal range (Fig. 8B). Moreover, CS-Cur-TGNCs exerted superior therapeutic effects on suppressing joint swelling and relieving joint inflammation. The clinical scores of the model rats increased rapidly, and the progression of GA was inhibited to different extents after treatment. Among these combinations, CS-Cur-TGNCs maximally inhibited GA disease progression, resulting in the lowest clinical score (Fig. 8C). Joint swelling was evaluated by measuring the circumference of the ankle joint. Similarly, CS-Cur-TGNCs significantly alleviated joint swelling and inflammation in GA rats (Fig. 8D). We measured the resting time, distance, and movement speed of GA rats in different treatment groups for 30 min. The results also showed that the CS-Cur-TGNCs treatment significantly promoted the recovery of mobility in GA rats (Fig. 8E-G). The biosafety of these nanopreparations is also considered in the treatment of this disease. After 30 days of continuous administration of the different preparations, vital organs, such as the heart, liver, spleen, lung and kidney, were observed via H&E staining. The results showed that there was no significant difference between the different groups (Fig. 8H), indicating that CS-Cur-TGNCs have a good safety profile. The blood of the rat was assayed for blood biochemistry and the results are shown in Fig. S4, which indicates that CS-Cur-TGNCs do not affect the normal functioning of the liver, kidneys and heart. Therefore, we concluded that CS-Cur-TGNCs are safe and effective for the treatment of arthritic diseases.

Fig. 8

CS-Cur-TGNCs Promotes joint recovery in GA. (A) Representative photos of the ankle joint of different groups of rats 24 h after MSU injection. (B) The changes of temperature in the ankle joints of different groups of rats. (C, D) Changes in treatment scores and ankle circumference in different groups of GA rats. (E-G) Time at rest, distance traveled, and speed of movement for 30 min in different groups of GA rats. (H) H&E staining of heart, liver, spleen, lung and kidney sections in each group. Data are shown as the mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001