A Study on the Treatment of Rheumatoid Arthritis Using a Novel GelMA-HAMA Dual-Network Hydrogel Microneedle Loaded with MTX-NCs in Combination with Adalimumab

TL;DR

A new hydrogel microneedle patch was developed to deliver drugs for rheumatoid arthritis, offering a painless and effective treatment option.

Contribution

A novel dual-network hydrogel microneedle patch loaded with methotrexate nanocrystals was developed for transdermal RA treatment.

Findings

The MTX-NCs had a spherical shape, average size of 325.72 nm, and 61.3% drug-loading capacity.

The microneedle patch showed high puncture efficiency and strong anti-inflammatory effects in vitro and in vivo.

Combination therapy with Adalimumab enhanced the anti-inflammatory efficacy and reduced joint damage in a rat RA model.

Abstract

This study developed a transdermal drug delivery system for Rheumatoid Arthritis (RA) using a dual-network hydrogel microneedle patch loaded with methotrexate nanocrystals (DHMN@MTX-NCs), and explored its synergistic therapy with Adalimumab (ADA) for a painless, long-acting, and targeted RA treatment. This study synthesized Methacrylated Hyaluronic Acid and Methacrylated Gelatin. MTX-NCs were prepared by solvent-antisolvent precipitation and incorporated into a dual-network hydrogel microneedle patch via centrifugal molding. Evaluations included pharmaceutical properties, mechanical strength, drug release, in vitro anti-inflammatory effects on RAW 264.7 cells, and therapeutic efficacy in a rat RA model. The experimental results show that the prepared MTX-NCs present a spherical shape, an average size of 325.72 nm, a PDI of 0.154, and a drug-loading capacity of 61.3%. The microneedle patch exhibited high puncture efficiency and suitable swelling. In vitro, DHMN@MTX-NCs combined with ADA most strongly inhibited macrophage migration, upregulated IL-10, and downregulated TNF-α, IL-1β, NO, iNOS, and COX-2. In vivo, both monotherapy and combination therapy reduced joint swelling, bone erosion, and histopathological damage. Ultimately, the study demonstrated the synergistic anti-inflammatory efficacy of DHMN@MTX-NCs combined with ADA, providing a novel, non-invasive, and targeted therapeutic strategy for RA.

Figures

Full text

1. Introduction

Rheumatoid Arthritis (RA) is a chronic autoimmune disorder characterized by high prevalence, recurrence, and disability rates, yet it is not an incurable malignancy [1,2]. Globally, the prevalence of rheumatoid arthritis (RA) is on the rise, with an estimated 17.6 million people affected worldwide, and the number of younger patients has also increased [3,4]. In China, the incidence of RA ranges from 0.26% to 0.5% [5], affecting individuals across all age groups. The precise etiology of RA remains unclear [6], although accumulating evidence indicates that it results from a complex interplay of genetic, immunological, infectious, and environmental factors [7]. Typical clinical manifestations include morning stiffness, joint swelling, pain, and deformities, and particularly in the hands, wrists, and ankles, along with subcutaneous nodules and osteoporosis [8,9]. Severe joint inflammation may lead to cardiac and pulmonary complications [10,11], and in advanced stages, can result inpermanent disability.

Current therapeutic strategies for RA include oral administration, intra-articular injection, and transdermal delivery systems. Oral medications typically require prolonged treatment cycles and are often associated with significant systemic side effects. Intra-articular injections, while effective, demand technical expertise, cause discomfort, are unsuitable for self-administration, and pose infection risks due to repeated invasive procedures. Transdermal delivery offers improved patient compliance and circumvents hepatic first-pass metabolism; however, the barrier function of the skin’s stratum corneum limits drug absorption, reducing efficacy.

There is a pressing need for targeted, effective RA therapies that ensure high drug utilization, minimal toxicity, and favorable patient adherence. Microneedles, as a novel transdermal system, enable painless, minimally invasive, and convenient drug administration with enhanced skin permeability, making them suitable for chronic and immune-mediated diseases [12]. Based on structural design, microneedles are classified into hollow, solid, coated, dissolving, and hydrogel microneedles (HMN). Among these, HMN exhibit superior mechanical strength and efficient stratum corneum penetration, enabling the delivery of large or non-lipophilic molecules and enhancing bioavailability. Upon insertion into the skin, they absorb interstitial fluid, swell, and establish sustained-release channels for controlled drug delivery [13]. Compared to other variants, HMN maintain these conduits for extended periods, rendering them especially promising for treating cardiovascular [14,15] and autoimmune diseases [16,17,18]. Drug release kinetics depend on the degree of cross-linking in the hydrogels network, which is dictated by the constituent materials. Single-network hydrogels often lack sufficient mechanical strength due to simple structures. Dual-network hydrogels improve upon this by combining a brittle and a tough network, enhancing strength, toughness, and structural stability. Their denser microstructure also reduces burst release and allows for tunable, sustained drug release profiles [19].

Natural polymers are widely used in hydrogels due to their biocompatibility, safety, and cell affinity. Commonly used polymers include gelatin, hyaluronic acid, sodium alginate, and chitosan. Gelatin (Gel), derived from partial hydrolysis animal collagen, exhibits outstanding biodegradability and biocompatibility. Hyaluronic Acid (HA), a linear polysaccharide and key component of the extracellular matrix and articular cartilage [20], plays critical roles in regulating cell proliferation, inflammation responses, and tissue regeneration. Chemical modification via methacrylation enhances the functional properties of both: GelMA demonstrates improved thermal stability, mechanical strength, and biocompatibility, serving as the primary network; HAMA contributes enhanced toughness and bioactivity, functioning as the secondary reinforcing network [21,22]. This complementary dual-network architecture not only significantly improves the hydrogel’s compressive modulus and puncture resistance but also increases its drug-loading capacity. Encapsulating MTX within this transdermal dual-network system circumvents the systemic toxicity associated with oral MTX administration including dose-limiting gastrointestinal disturbances, hepatotoxicity, and myelosuppression. Moreover, advances in RA pathophysiology have identified multiple druggable targets now entering clinical evaluation, such as TNF-α, B cells, T cells, monocyte–macrophages, and synovial fibroblasts [23,24].The first biological product used in clinical practice for the treatment of RA is the TNF-a inhibitor. The representative drug ADA is a fully humanized monoclonal antibody biological product that can specifically bind to TNF-a and block its interaction with p55 and p75 cell surface TNF receptors, thereby counteracting the biological function of TNF. In the clinical treatment of RA, when a biological disease-modifying antirheumatic drug or a targeted synthetic disease-modifying antirheumatic drug fails to achieve the treatment target, a combination therapy regimen can be considered as a priority. Clinical application has found that the use of ADA in combination with low-dose MTX can enhance the clinical response of RA patients, reduce antibody formation, and has significant therapeutic effects and medication safety.

This study aims to develop a novel dual-network hydrogel microneedle (DHMN) based on GelMA-HAMA to overcome existing limitations in transdermal RA therapy, enhance carrier stability, and provide an innovative strategy for effective, patient-friendly treatment.

2. Results

2.1. Synthesis and Characterization of HAMA, GelMA, and MTX-NCs

2.1.1. Synthesis and Characterization of HAMA and GelMA

As shown in Figure 1a, the FTIR spectrum of HA exhibits characteristic absorption peaks: the C–H stretching vibration at 2940 cm^−1^, the asymmetric carboxylate (COO^−^) stretching vibration at 1610 cm^−1^, and the C–O–C stretching vibration at 1150 cm^−1^. For methacrylic anhydride (MA), the anhydride functional group displays a diagnostic doublet corresponding to asymmetric and symmetric carbonyl (C = O) stretching vibrations at 1820 cm^−1^ and 1760 cm^−1^, respectively. In the spectrum of HAMA, all major HA characteristic peaks are retained. Notably, the C–H stretching peak undergoes a slight shift due to the chemical reaction with MA, and a new absorption band appears at 1635 cm^−1^, assigned to the alkene C = C stretching vibration. This peak corresponds to the carbon-carbon double bond introduced by the methacryloyl group, confirming the successful synthesis of HAMA. Figure 1b presents the FTIR spectrum of gelatin (Gel), showing typical amide bands: the N–H stretching vibration at 3280 cm^−1^ and the C=O stretching vibration at 1630 cm^−1^ (amide I). The characteristic anhydride doublet of MA is also observed at 1820 cm^−1^ and 1760 cm^−1^. In the GelMA spectrum, the amide I and N–H stretching peaks exhibit minor shifts, indicating structural modification. Furthermore, a distinct C = C stretching vibration peak emerges at 1635 cm^−1^, which is attributed to the vinyl group of the grafted methacryloyl moiety. This provides clear evidence for the successful functionalization of gelatin with methacrylic anhydride, confirming the formation of GelMA.

The formation of the C = C bonds via esterification of HA and Gel with MA enables subsequent UV-initiated crosslinking to form polymer networks. Further confirmation of chemical modification was obtained through ^1^H NMR spectroscopy. As shown in Figure 1c, the ^1^H NMR spectrum of HA displays the methyl proton signal of the acetamido group (–COCH_3_) at 1.9 ppm and sugar ring protons between 3.3 and 3.8 ppm. In the HAMA spectrum, two new resonance signals appear at 6.1 ppm and 5.7 ppm, the characteristic vinyl protons (=CH_2_) of the methacryloyl group appear at 6.1 ppm and 5.7 ppm, corresponding to the vinyl protons (=CH_2_) of the methacryloyl group, this serves as a key signature of successful derivatization [25]. The methyl proton peak (–CH_3_) of the methacryloyl group is observed at 2.0 ppm, while the acetamido methyl peak of HA remains unchanged at 1.9 ppm, consistent with the parent HA spectrum. These findings are in agreement with previous reports, confirming the successful synthesis of HAMA. The degree of substitution was calculated to be approximately 20%, based on integration analysis using Formula (1). In Figure 1d, native gelatin shows characteristic aliphatic protons at 0.8 ppm, the β-methylene protons at 1.9 ppm, and the α-protons at 3.2 ppm. Upon modification, the GelMA spectrum reveals the appearance of the vinyl proton signals (=CH_2_) at 6.1 ppm and 5.7 ppm, unambiguously confirming covalent attachment of the methacryloyl groups [26]. All principal proton resonances of gelatin are preserved in GelMA, indicating that the backbone structure remained intact during functionalization. These results are consistent with prior studies, verifying the successful preparation of GelMA, with a degree of substitution of approximately 60%, determined by Formula (2).

2.1.2. Synthesis and Characterization of MTX-NCs

Morphological examination by TEM (Figure 2a) revealed that the methotrexate nanocrystals (MTX-NCs) were spherical and uniform in size. Dynamic light scattering (DLS) analysis (Figure 2b,c) showed that the MTX-NCs had an average hydrodynamic diameter of 325.72 ± 5.7 nm, a polydispersity index (PDI) of 0.154 ± 0.016, and a zeta potential of −11.4 ± 1.2 mV. FTIR spectroscopy was used to evaluate potential chemical interactions (Figure 2d). The spectrum of the physical mixture closely resembled a superposition of the individual spectra of methotrexate and excipients, confirming simple physical blending without interaction. In contrast, the spectrum of the MTX-NCs showed the disappearance of characteristic MTX absorption bands, specifically the C=O stretching vibration (carboxylic acid) at 1710 cm^−1^ and the N–H stretching (amide) at 3218 cm^−1^. No new peaks or significant shifts in existing peaks were observed, suggesting no chemical interactions occurred during nanocrystal formation. Additionally, the absence of characteristic anhydrous ethanol signals confirmed its complete removal during preparation.

Thermal properties were evaluated by DSC (Figure 2e). Pure MTX exhibited a sharp endothermic melting peak at 155.73 °C [27], while mannitol (MNT) displayed a distinct endothermic transition at 167.72 °C. PVP K30 displayed a broad, featureless thermogram with no discernible melting event. The physical mixture retained the characteristic endothermic peaks of both MTX and MNT, indicating the coexistence of crystalline components. Notably, the MTX melting peak was absent in the MTX-NCs, indicating that the drug was converted into an amorphous state during processing. This observation was further supported by powder X-ray diffraction PXRD analysis (Figure 2f). The characteristic crystalline diffraction peaks of MTX were largely absent in the MTX-NCs pattern, whereas residual peaks corresponded to crystalline MNT. These findings confirm that MTX exists in an amorphous form within the nanocrystal formulation, consistent with DSC results. Collectively, these data demonstrate that the preparation process disrupted the crystalline lattice of MTX, leading to its amorphous dispersion within the NCs. The drug-loading capacity of MTX-NCs is summarized in Table 1.

2.2. Preparation and Characterization of DHMN@MTX-NCs

Figure 3a illustrates the fabrication process of DHMN@MTX-NCs. As shown in Figure 3b–d, each hydrogel group exhibits an irregular three-dimensional porous structure with well-interconnected pores. The dual-network hydrogel formed by blending HAMA and GelMA (Figure 3d) displays a smooth surface without granular morphology. Its internal porosity is significantly reduced, accompanied by a higher density of micropores and more uniform pore size distribution. This indicates the formation of an interpenetrating network structure between HAMA and GelMA, which can better balance force distribution and effectively enhance the mechanical strength of the hydrogel [28]. This improved strength meets the subsequent experimental requirements for hydrogel microneedle (MN) performance.

Figure 3e shows that the FTIR spectrum of the physical mixture is a simple superposition of the spectra from MTX and the blank dual-network hydrogel MNs, confirming simple physical mixing. In contrast, the FTIR spectrum of DHMN@MTX-NCs closely resembles that of the blank dual-network MNs, with the characteristic absorption peaks of MTX absent and no new peaks observed. This indicates successful encapsulation of MTX-NCs within the dual-network hydrogel matrix without chemical interaction, likely mediated by hydrogen bonding. Similarly, the DSC thermogram (Figure 3f) shows an endothermic melting peak of MTX near 155 °C in the physical mixture, whereas this peak is absent in DHMN@MTX-NCs. This further confirms the effective loading of MTX-NCs and successful preparation of DHMN@MTX-NCs.

Optical microscopy (Figure 4a) reveals that the microneedles possess a pyramid-shaped structure free of surface bubbles. FESEM imaging (Figure 4b) confirms that DHMN@MTX-NCs maintain a complete array of sharp-tipped pyramidal microneedles, with no visible bending or breakage, consistent with the morphology of blank dual-network MNs (DNMN). Mechanical testing (Figure 4c,d) demonstrates that under a 410 g load, the microneedle height was compressed from 800 µm to 614 ± 1.8 µm, corresponding to a compression rate of approximately 8.1 ± 2.4%. This indicates good structural toughness, suitable for subsequent experimental use. The photo-initiated crosslinking method employed ensures stable network formation while minimizing potential toxicity associated with conventional chemical crosslinkers.

The release of Rhodamine B from three different MN formulations was evaluated (Figure 4e). At both 25 °C and 37 °C, the dual-network MNs exhibited significantly faster and greater diffusion of Rhodamine B compared to single-network GelMA or HAMA MNs, with cumulative release increasing over time. Enhanced release was observed at elevated temperature at all time points. The pronounced swelling of the dual-network MNs during release further accelerated the release rate and extent, indicating that the dual-network architecture can effectively modulate drug release kinetics.

Low moisture uptake is critical for DHMN to maintain mechanical integrity during skin insertion and drug delivery [29]. As shown in Figure 5a–c, under three different relative humidity conditions (33%, 65%, and 96% RH), both DNMN and DHMN@MTX-NCs exhibited low hygroscopicity, with moisture absorption remaining below 15%. The dynamic swelling behavior is presented in Figure 5d. Swelling equilibrium was reached at 120 min, with swelling ratios of 523.31 ± 14.18% for DNMN and 466.19 ± 13.67% for DHMN@MTX-NCs. The slightly lower swelling ratio of DHMN@MTX-NCs is likely attributable to the presence of loaded MTX-NCs perturbing the crosslinked network structure; however, it remains within an acceptable range for subsequent applications. Mechanical functionality was further assessed by penetration tests through four layers of Parafilm^®^ M (Table 2). DHMN@MTX-NCs achieved a penetration rate of 97.8 ± 1.47% across five repeated trials, fulfilling the experimental criteria.

2.3. In Vitro Release Study Results of DHMN@MTX-NCs

As shown in Table 3, the cumulative release of free MTX and DHMN@MTX-NCs over a 72 h period was 43.69 ± 2.06% and 91.84 ± 1.62%, respectively. At the 72 h time point, the cumulative release from DHMN@MTX-NCs was 2.10-fold higher than that of free MTX. This demonstrates that the dual-network hydrogel microneedles significantly enhanced MTX releas, effectively overcoming its inherently low aqueous solubility.

Furthermore, as shown in Table 4, the in vitro cumulative release profiles of both MTX and DHMN@MTX-NCs conform to the first-order kinetic model.

2.4. In Vitro Transdermal Permeation Study Results of DHMN@MTX-NCs

As shown in Table 5, the cumulative transdermal permeation of MTX, MTX-NCs, and DHMN@MTX-NCs over 72 h was 25.43 ± 2.16%, 43.77 ± 1.94%, and 87.86 ± 1.34%, respectively. The permeation achieved with DHMN@MTX-NCs was 3.10-fold higher than that of free MTX, indicating that the dual-network hydrogel microneedles effectively penetrated the stratum corneum and significantly enhanced MTX transdermal delivery.

Furthermore, as shown in Table 6, the in vitro cumulative transdermal permeation profiles of MTX, MTX-NCs, and DHMN@MTX-NCs all conform to the first-order kinetic model.

2.5. In Vitro Study of DHMN@MTX-NCs

2.5.1. Hemolysis Assay Results

As shown in Figure 6, at a concentration of 100 mg·mL^−1^, the hemolysis rates for the DHMN and DHMN@MTX-NCs groups were 1.06% and 1.63%, respectively. These values are well below the ISO 10993-5 safety threshold of <5% [30], indicating minimal hemolytic potential of both the excipient carrier and the drug formulation on red blood cell membranes. Following UV irradiation, the hemolysis rates increased slightly across all groups but remained below 2%. This confirms that the excipient carrier exhibits excellent hemocompatibility both before and after UV crosslinking, providing a solid safety foundation for subsequent animal studies with DHMN@MTX-NCs.

2.5.2. Cytotoxicity Results

RAW 264.7 is a mouse macrophage cell line induced by Abelson Murine Leukemia Virus (A-MuLV), derived from tumor tissue of male BALB/c mice, and widely used in research on inflammation, immunity, and osteoclasts.

Figure 7a shows that the DHMN group, used as the carrier material, exhibited no significant cytotoxicity toward either normal or polarized RAW 264.7 cells, demonstrating good biocompatibility. For the MTX and DHMN@MTX-NCs groups, within the concentration range of 4.54 to 13.63 mg·mL^−1^, cell viability remained above 80% despite increasing MTX concentrations, indicating acceptable safety under these conditions. However, at higher concentrations (18.18 to 22.72 mg·mL^−1^), cell viability dropped below 80% in both groups.

In the combination group, when ADA concentration was maintained between 0.37 mg·mL^−1^ and 0.74 mg·mL^−1^, cell viability gradually decreased with increasing MTX concentration but remained above 80%, suggesting favorable biosafety. At a higher ADA concentration of 7.4 mg·mL^−1^, further increases in MTX concentration reduced the viability of both normal and polarized RAW 264.7 cells below 80%, indicating dose-dependent cytotoxic effects.

2.5.3. Live/Dead Cell Staining Results

Fluorescent staining using Calcein-AM labels viable cells green, enabling assessment of cell viability based on green fluorescence intensity. As shown in Figure 7b,c, the PBS control group displayed abundant green-fluorescent cells, with a viability of 97%, confirming high biosafety. In the DHMN@MTX-NCs group, green fluorescence intensity decreased with increasing drug concentration, resulting in a viability of 89%. In the combination therapy group, a marked increase in cell death was observed; as methotrexate (MTX) concentration increased, viability declined to a minimum of 84%. These results indicate that combination therapy exerts a stronger biological effect than microneedle treatment alone, consistent with the findings from the cytotoxicity assay.

2.5.4. Scratch Wound Healing Assay Results

As shown in Figure 7d, polarized RAW 264.7 cells retained intrinsic migratory capacity. In the PBS group, robust cell migration was observed, with a migration rate of approximately 40% at 24 h. The DHMN@MTX-NCs group showed a reduced migration rate of about 30% at 24 h, indicating moderate inhibition of cell motility. The combination group exhibited more pronounced inhibition, with a migration rate of only ~17% at 24 h. Furthermore, the inhibitory effect in the combination group was concentration-dependent: higher drug concentrations led to greater suppression of migration in polarized RAW 264.7 cells.

2.5.5. Assessment of Anti-Inflammatory Activity

IL-10 is an anti-inflammatory cytokine, whereas TNF-α and IL-1β are pro-inflammatory cytokines. Nitric oxide (NO) production serves as a key indicator of successful polarization of RAW 264.7 cells into a pro-inflammatory phenotype.

As shown in Figure 8, a highly significant difference (p < 0.001) was observed between the blank control group and the model group. This confirms a marked reduction in the anti-inflammatory cytokine IL-10 in the model group, validating successful establishment of the inflammatory model. Both DHMN@MTX-NCs (at all tested doses) and the combination group demonstrated significant anti-inflammatory activity, effectively upregulating IL-10 and downregulating pro-inflammatory mediators (IL-1β, TNF-α) and NO levels. Notably, the anti-inflammatory effect of the combination therapy was superior to that of either DHMN@MTX-NCs or the positive control drug alone.

In summary, DHMN@MTX-NCs exhibit substantial anti-inflammatory efficacy, and their combination with ADA produces an enhanced therapeutic effect, supporting the rationale for subsequent in vivo experiments.

2.5.6. Western Blot Analysis

The activation of inducible nitric oxide synthase (iNOS) promotes the production of nitric oxide (NO), a key pro-inflammatory mediator that stimulates the release of cytokines such as TNF-α, IL-1β, and IL-10 [31]. Similarly, the enzymatic products of cyclooxygenase-2 (COX-2) also contribute to the release of NO and other pro-inflammatory factors [32]. Elevated levels of NO and pro-inflammatory mediators in vivo can induce DNA damage in tissues, exacerbate inflammatory cell infiltration and edema, and accelerate the progression of chronic inflammatory diseases [33].

As shown in Figure 9a–c, a significant difference (p < 0.01) was observed between the model group and the blank control group, confirming successful induction of the inflammatory model. Both the positive control group and the microneedle (MN) group exhibited significant downregulation of iNOS and COX-2 protein expression compared to the model group (p < 0.01), indicating effective anti-inflammatory activity. Notably, the MN group showed a stronger inhibitory effect on iNOS expression than the positive control group. Moreover, a significant difference (p < 0.01) was observed between the MN group and the combination therapy group, with the latter showing greater suppression of both iNOS and COX-2 protein levels. These findings further support that the co-administration of DHMN@MTX-NCs and ADA exerts a synergistic anti-inflammatory effect.

2.6. In Vivo Study of DHMN@MTX-NCs

2.6.1. Results of Skin Penetration, Healing, and Irritation in Rats Treated with DHMN@MTX-NCs

As shown in the left panel of Figure 10a, the skin penetration rate of DHMN@MTX-NCs exceeded 90%, demonstrating that the microneedles possess sufficient mechanical strength to penetrate the stratum corneum and create microchannels for drug delivery. No signs of irritation such as erythema, swelling, or blistering were observed on rat skin, indicating the formulation is well-tolerated and non-irritating.

Furthermore, Figure 10a,b show that skin treated with DHMN@MTX-NCs healed completely within approximately 30 min. The micropores created by the microneedles were fully closed after this period, with no residual marks, redness, or blister formation. These observations confirm the rapid skin recovery and excellent biocompatibility of DHMN@MTX-NCs.

2.6.2. In Vivo Skin Swelling Results of DHMN@MTX-NCs

As shown in Figure 10c, the FESEM image clearly reveals significant swelling of DHMN@MTX-NCs after insertion into rat skin compared to their pre-insertion state. The needle height increased from 800 μm to 826 ± 4.5 μm, indicating that the microneedles absorbed interstitial fluid from the skin tissue and swelled following penetration, thereby facilitating drug release.

2.6.3. Skin Histology in Rats Treated with DHMN@MTX-NCs

As shown in Figure 10d, DHMN@MTX-NCs formed needle-shaped cavities with smooth contours in the rat skin. The length of these microchannels was approximately 780 μm, close to the original height of the fabricated microneedles (800 μm). This indicates successful penetration of the stratum corneum, followed by in situ swelling within the skin tissue and formation of stable conduits suitable for sustained drug delivery.

2.6.4. Body Weight Changes in Rats

The experimental timeline for the induction of adjuvant-induced arthritis (AA) in rats and the corresponding treatment regimen is schematically illustrated in Figure 11a. As shown in Figure 11b, a general decline in body weight was observed across all groups starting from day 8. Following the initiation of intervention on day 15, rats in the blank control group exhibited a steady increase in body weight. In contrast, the model group continued to lose weight until a slow recovery began around day 28. Earlier recovery was observed in the treatment groups: on day 24 for the positive control group, day 20 for the ADA group, and as early as day 16 for both the DHMN@MTX-NCs and combination groups.

In summary, all treatment groups showed statistically significant differences compared to the model group at each monitored time point (p < 0.001), indicating that all interventions promoted body weight recovery to varying degrees. Notably, by the end of the treatment period, the body weight of rats in the combination group approached that of the blank control group.

2.6.5. Arthritis Score Changes in Rats

As shown in Figure 11c, arthritis scores in the blank control group remained at 0 throughout the study. In contrast, scores in the model group increased continuously from day 8 to day 40, followed by a slight decline after day 40. Compared to the model group, the positive control group exhibited a gradual rise in scores from day 8 to day 28, after which scores began to decrease, indicating progressive alleviation of paw swelling. The DHMN@MTX-NCs group showed a reduction in arthritis scores starting from day 24, reflecting decreasing inflammation, while the ADA group demonstrated improvement beginning on day 20. Notably, the combination group exhibited the earliest therapeutic response, with scores declining from day 16.

Overall, during the treatment period from day 8 to day 48, all treatment groups differed significantly from the model group (p < 0.001), confirming that each treatment effectively ameliorated arthritis-related swelling. Moreover, both the DHMN@MTX-NCs and combination groups achieved greater improvements in arthritis scores than the positive control group.

2.6.6. Paw Swelling Changes in Rats

As shown in Figure 11d, the paw volume (measured by water displacement) of rats in the blank control group remained relatively constant across all monitored time points. In contrast, paw volume in the model group increased gradually from day 8 to day 40, followed by a slight decrease thereafter, likely due to mild self-limiting inflammation mediated by the autoimmune response. The positive control group began to show a reduction in paw volume on day 36, indicating partial mitigation of inflammatory swelling. The DHMN@MTX-NCs group exhibited an earlier therapeutic response, with paw volume decreasing starting from day 24, suggesting superior anti-inflammatory efficacy compared to the positive control. Both the ADA group and the combination group showed a decline in paw volume beginning on day 20. Notably, the combination group demonstrated a more pronounced reduction and a faster recovery rate than the ADA group alone. These results indicate that both DHMN@MTX-NCs and the combination therapy effectively alleviate paw swelling in arthritic rats.

2.6.7. Paw Thickness Changes in Rats

As shown in Figure 11e, paw thickness in the blank control group remained stable throughout the treatment period. In contrast, the model group exhibited progressive thickening from day 8, peaking around day 44, followed by a slight decline possibly attributable to mild spontaneous resolution of inflammation via immune regulation. A reduction in paw thickness, indicative of inflammation alleviation, was first observed on day 36 in the positive control group. The DHMN@MTX-NCs group showed an earlier and more substantial decrease beginning on day 24, outperforming the positive control. Both the ADA and combination groups exhibited reduced paw thickness starting from day 20, with the combination group displaying a greater magnitude of reduction and accelerated recovery compared to ADA monotherapy.

In summary, all treatment groups differed significantly from the model group in terms of paw thickness reduction (p < 0.001), confirming that each intervention induced a marked improvement in joint pathology.

2.6.8. X-Ray Findings

Representative X-ray images of rat paws are shown in Figure 12. Rats in the blank control group displayed no signs of swelling, with intact bone architecture and no evidence of structural damage. In contrast, the model group exhibited marked paw swelling, narrowed joint spaces, evident bone erosion, osteoproliferation, and mild limb stiffness. The positive control group showed partial restoration of joint space compared to the model group, although significant soft tissue swelling persisted. Both the ADA and DHMN@MTX-NCs groups showed noticeable reductions in paw swelling and improved joint clarity relative to the model group. Notably, the combination group exhibited the most pronounced radiographic improvement, with markedly reduced swelling and well-preserved, clearly defined joint structures. These imaging findings are consistent with the results from paw thickness measurements and arthritis scoring, collectively supporting the therapeutic efficacy of the treatments.

2.6.9. Analysis of Rat Organ Indices

The spleen and thymus are primary immune organs that are susceptible to inflammatory damage under pathological conditions, often leading to organ enlargement and increased organ indices. As shown in Figure 13a,b, the spleen and thymus indices in the blank control group were low, reflecting normal physiological function. In contrast, both indices were significantly elevated in the model group compared to the control, indicating systemic inflammation-induced immune organ impairment. Relative to the model group, the positive control group exhibited a modest reduction in these indices. Further decreases were observed in both the DHMN@MTX-NCs and ADA groups, with the combination group showing the most pronounced lowering effect. In summary, all treatment groups significantly reduced spleen and thymus indices compared to the model group (p < 0.001), with effects superior to those of the positive control. These findings demonstrate that the interventions effectively alleviated systemic inflammation and restored immune homeostasis, highlighting their substantial therapeutic potential.

2.6.10. Histopathological Analysis of Ankle Joint Tissues

H&E staining results (Figure 14a) revealed distinct histopathological features across experimental groups. The blank control group displayed intact joint architecture, an orderly synovial lining, and smooth articular cartilage surfaces, with no evidence of inflammatory cell infiltration or structural damage. In contrast, the model group exhibited severe arthritic changes, including extensive inflammatory cell infiltration (predominantly lymphocytes and neutrophils), marked synovial hyperplasia and thickening with pannus formation, and erosion of cartilage and subchondral bone, resulting in irregular cartilage surfaces and narrowed joint spaces. Treatment with the positive control drug markedly ameliorated these pathological changes, reducing inflammatory infiltration and synovial overgrowth while partially preserving cartilage integrity. Both the DHMN@MTX-NCs and ADA groups demonstrated clear chondroprotective effects, characterized by significantly reduced inflammatory cell influx into the joint cavity, attenuated synovial hyperplasia, and relatively preserved cartilage structure, with comparable degrees of improvement. Notably, the combination group showed the most robust joint protection, with joint architecture nearly restored to normal, minimal residual inflammation, almost complete suppression of synovial hyperplasia, and a smooth, intact cartilage surface.

Safranin O/Fast green staining results (Figure 14b) further corroborated these observations, as the intensity of red Safranin O staining correlates directly with proteoglycan content and cartilage matrix health. The blank control group exhibited intense and uniform red staining, indicative of a rich, healthy extracellular matrix. In contrast, the model group showed markedly diminished or absent staining, reflecting severe proteoglycan depletion and matrix degradation. Partial restoration of staining intensity was observed in the positive control, DHMN@MTX-NCs, and ADA groups, suggesting a protective effect against matrix breakdown. The combination group displayed the strongest and most homogeneous red staining, closely resembling the blank control, demonstrating superior efficacy in preserving and restoring cartilage matrix components.

Collectively, these histopathological findings indicate that DHMN@MTX-NCs, ADA, and the positive control drug all effectively mitigate joint inflammation and protect against cartilage destruction in AA rats. Among the treatments, the combination of DHMN@MTX-NCs and ADA produced a synergistic therapeutic effect, achieving near-complete suppression of disease progression and restoration of joint morphology to a near-normal state.

2.6.11. Immunohistochemical Analysis of Ankle Joints

Bone erosion, a hallmark of rheumatoid arthritis, arises from excessive osteoclast activation coupled with impaired osteoblast function [34]. TNF-α promotes aberrant osteoclastogenesis, stimulates osteoclast differentiation [35], and induces osteoblast apoptosis [36]. IL-1β inhibits osteoblast proliferation and migration and acts synergistically with TNF-α to enhance osteoclast differentiation [37]. Similarly, IL-6 contributes to osteoclast formation [38] and suppresses osteoblast differentiation [39].

Immunohistochemical analysis of rat ankle joints is presented in Figure 14c–f. Semi-quantitative assessment of IL-1β expression revealed a statistically significant increase in the model group compared to the blank control (p < 0.001), visually confirmed by extensive brown-yellow staining, validating successful induction of the AA model. Both the DHMN@MTX-NCs and combination groups showed significant downregulation of IL-1β relative to the model group (p < 0.001), accompanied by markedly reduced positive staining. Furthermore, both groups outperformed the positive control in suppressing IL-1β expression (p < 0.001), as evidenced by weaker immunoreactivity.

IL-6 expression followed a similar trend: levels were significantly elevated in the model group versus the control (p < 0.001), with abundant positive staining. Treatment with DHMN@MTX-NCs or the combination therapy significantly reduced IL-6 expression compared to the model group (p < 0.001). Both groups also demonstrated greater efficacy than the positive control (p < 0.001). A direct comparison between the DHMN@MTX-NCs and combination groups revealed that the latter achieved a more pronounced reduction in IL-6 (p < 0.001), indicating enhanced therapeutic benefit through combination intervention at the same drug concentration.

For TNF-α expression, a significant upregulation was observed in the model group relative to the blank control (p < 0.001). Both DHMN@MTX-NCs and the combination therapy effectively suppressed TNF-α levels compared to the model group (p < 0.001). Moreover, both treatments exhibited stronger inhibition than the positive control (p < 0.001), as reflected by reduced immunostaining intensity.

3. Discussion

In this study, the hydrogel components HAMA and GelMA were successfully synthesized. FESEM observations revealed that both polymers exhibit characteristic three-dimensional porous architectures. When blended to construct a double-network hydrogel, the internal pore volume significantly decreased, the micropore density increased, and the pore size distribution became more uniform. These morphological changes strongly support the formation of a dense and fully cross-linked interpenetrating polymer network (IPN) between HAMA and GelMA. This IPN structure not only enables uniform spatial stress dispersion, significantly enhancing the material’s compressive modulus and resistance to plastic deformation, but also establishes a nanoscale-confined, high-surface-area microenvironment ideal for efficient drug loading, confined storage, and sustained molecular release.

Given that MTX exhibits extremely low aqueous solubility and lipophilicity, direct incorporation into the dual-network hydrogel microneedles inevitably leads to drug crystallization, heterogeneous distribution, and in vitro release retardation. Accordingly, this study employed a nanocrystal formulation strategy to enhance MTX’s apparent solubility via particle size reduction. In vitro cumulative release studies demonstrated that MTX-NCs achieved a 1.59-fold higher release at 72 h compared with free MTX, accompanied by a significantly accelerated initial release rate. Furthermore, the release profiles of both MTX and MTX-NCs fitted the first-order kinetic model, yet their rate constants and terminal cumulative release amounts differed significantly. The markedly lower k and Q_n_ values of free MTX were attributable to its coarse crystalline morphology and intrinsic dissolution-limited behavior rather than poor “solubility” per se, where dissolution kinetics govern the overall release rate. This study did not quantify the drug residues in the skin tissue, which is a key piece of data that needs to be supplemented in future research. MTX, as a dihydrofolate reductase inhibitor, can profoundly alter the cellular metabolic state by interfering with nucleic acid synthesis and affecting one-carbon metabolism. Compared with free large-sized MTX crystals, MTX nanocrystals, due to their smaller size and larger specific surface area, may be more easily taken up by macrophages through endocytosis, thereby achieving higher intracellular drug concentrations, more effectively interfering with their metabolic pathways, and thus more likely to lead to M2-type polarization.

In the swelling performance characterization experiment, PBS solution was used to simulate the interstitial fluid of the skin. Only the water absorption and swelling performance of the hydrogel was used to evaluate the microneedles, and the volume of the PBS solution was selected to exceed that of the interstitial fluid of the skin. Therefore, the water gel microneedles swelled severely, and the needle body became hemispherical. No swelling disintegration or structural collapse was observed, confirming that they have excellent resistance to excessive hydration and mechanical stability. This property ensures that after the microneedles are inserted into the skin, they can rapidly form a through microchannel driven by tissue fluid, promoting the directional delivery of drugs from the backing layer to the deep dermis. At the same time, the fully swollen microneedles can still be completely and undamagedly removed, completely avoiding the common risk of fracture and residue of traditional insoluble microneedles, and eliminating the safety hazard of local inflammatory reactions caused by foreign body retention from the source.

This study confirmed that DHMN@MTX-NCs can effectively inhibit cell migration in a concentration-dependent manner. This phenomenon is related to the effects of MTX in regulating the cytoskeleton and inhibiting the production of matrix metalloproteinases (MMPs) [40]. The local high concentration and sustained-release characteristics achieved through microneedles may provide ideal conditions for continuously inhibiting MMP activity and interfering with integrin signaling, thereby preventing the invasion of synovial cells into cartilage. This provides a cellular-level explanation for the observed protection of joint structure. The DHMN@MTX-NCs group had a superior ability to inhibit cell migration compared to the positive control group, and the combination group showed the best inhibitory effect, with the inhibitory ability of cell migration presenting a concentration-dependent manner. Its ability to regulate the levels of inflammatory factors and down-regulate proteins was superior to that of the positive control group, and the effect was more significant when combined with ADA. In the RA synovium, M1-type macrophages are overly activated and secrete large amounts of cytokines such as TNF-α and IL-6, driving inflammation and bone destruction; while M2-type macrophages play a role in suppressing inflammation and promoting tissue repair. The traditional pulsed administration of free MTX may lead to significant fluctuations in local drug concentration, which is insufficient to continuously and stably regulate the complex polarization process. The reason why the DHMN@MTX-NCs group and the combination group had better effects was that MTX in the microneedles could be slowly and continuously released as the hydrogel degraded, maintaining a stable and effective drug concentration locally, avoiding the concentration fluctuations caused by pulsed drug delivery, and thus achieving a more sustained and stable anti-inflammatory effect. The combination group had a better effect than the DHMN@MTX-NCs group alone because MTX and ADA exerted a synergistic effect: MTX inhibited the proliferation of activated T lymphocytes and reduced the cells that produced TNF-α; ADA specifically neutralized the already produced TNF-α and blocked its downstream inflammatory signals. The results of this study show that the sustained and slow local release of methotrexate mediated by microneedles significantly downregulated the protein expressions of iNOS and COX-2, and its effect was superior to that of the traditional positive drug administration method. The combined administration group demonstrated a stronger inhibitory effect. iNOS and COX-2 are the convergence effect proteins of two core pathways in the inflammatory response. iNOS is regulated by the NF-κB and JAK-STAT pathways, catalyzing the production of a large amount of nitric oxide, mediating oxidative stress and tissue damage. COX-2 is induced by the MAPK/AP-1 and NF-κB pathways, catalyzing the synthesis of prostaglandin E2, amplifying the inflammatory cascade reaction. In this study, the microneedle group’s ability to downregulate iNOS was significantly better than that of the positive group, suggesting that the sustained and slow local drug release may maintain the effective drug concentration at the lesion site, prolong the activation time of the adenosine signaling pathway, and thus more persistently block the binding activity of NF-κB to DNA. The experimental data show that the microneedle group’s ability to downregulate iNOS was better than that of the positive group, but although there was a statistical difference in the downregulation of COX-2, its effect magnitude seemed to be less than that of iNOS. This phenomenon may be due to the difference in the sensitivity of the two pathways to the continuous stimulation of methotrexate. The promoter region of iNOS contains multiple NF-κB binding sites, and its expression is highly dependent on the continuous activation of this transcription factor. COX-2, in addition to being regulated by NF-κB, is also redundantly regulated by multiple pathways such as MAPK/AP-1 and C/EBPβ. The efficient regulation of the iNOS/COX-2 signaling axis by the sustained and slow local release of methotrexate has dual potential advantages in the treatment of chronic inflammatory diseases. First, by targeting the inhibition of iNOS, it can reduce the generation of peroxynitrite and protect local tissues from peroxidation damage. Second, the long-term inhibition of COX-2 helps to reduce pain and vascular permeability mediated by prostaglandins while avoiding the gastrointestinal side effects caused by the non-selective inhibition of COX-1 by traditional NSAIDs.

In the rat skin swelling experiment, DHMN@MTX-NCs were easily stripped and did not easily break. Further, their efficacy in alleviating rheumatoid arthritis (RA) was evaluated in an adjuvant arthritis (AA) rat model. Compared with the commonly used clinical drug, diclofenac sodium cream, DHMN@MTX-NCs could more effectively slow down the progression of RA and reduce joint inflammation. Moreover, the combined administration group had a better therapeutic effect than the DHMN@MTX-NCs alone group. This administration strategy not only follows the clinical treatment strategy of combining methotrexate (MTX) with adalimumab (ADA), but also significantly enhances the bioavailability of the drug at the target site through the sustained release of MTX mediated by microneedles. The DHMN@MTX-NCs microneedle system developed in this study demonstrated remarkable anti-inflammatory and joint-protective effects in the AA rat model. The pathological core of RA involves immune-inflammatory dysregulation, synovial hyperplasia, and progressive cartilage and bone destruction, driven by a series of interwoven signaling pathways and intracellular processes [41]. This study observed that DHMN@MTX-NCs could significantly down-regulate TNF-α and IL-6 levels, which is attributed to its unique sustained-release kinetics. Compared with traditional topical preparations or pulsed administration, the local and continuous drug release provided by the hydrogel microneedles can more stably inhibit the continuous activation of these key signaling pathways. MTX, through its immunomodulatory effect, and ADA, through its specific neutralization of TNF-α, work synergistically, interrupting the inflammatory positive feedback loop from both inhibiting production and neutralizing clearance, which is consistent with the clinical logic that combination therapy is superior to monotherapy.

In summary, the DHMN@MTX-NCs microneedle system is not only an innovation in delivery technology. Its sustained-release, high-efficiency, and local-targeting characteristics enable it to deliver treatment in a manner that better aligns with the chronic pathophysiological features of RA. This study not only successfully overcame the multiple challenges of transdermal delivery of the hydrophobic drug MTX through reasonable material design and formulation techniques, but also established a new microneedle paradigm that can achieve local sustained release and precise combination therapy, providing a solid experimental basis and new technical ideas for the development of minimally invasive, efficient and safe treatment strategies for chronic inflammatory diseases such as rheumatoid arthritis.

4. Materials and Methods

4.1. Materials and Reagents

Hyaluronic Acid (HA), N,N-Dimethylformamide (DMF), Methacrylic Anhydride (MA), Gelatin (Gel), Methotrexate (MTX), Povidone K30 (PVP K30), and Adalimumab (ADA) were purchased from Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). Sodium Hydroxide (NaOH) was purchased from Merck KGaA (Darmstadt, Germany). Photoinitiator I2959 was purchased from BASF SE (Ludwigshafen, Germany). Absolute Ethanol (analytical grade), Sodium Bicarbonate (NaHCO_3_), and Mannitol (MNT) were purchased from Tianjin Komiou Chemical Reagent Co., Ltd. (Tianjin, China). Phosphate-Buffered Saline (PBS) (dry powder) was purchased from Hangzhou Shapu Biotechnology Co., Ltd. (Hangzhou, China).

Lipopolysaccharide (LPS, Catalog Number: L2880) was purchased from Beijing BioTope Biotechnology Co., Ltd. (Beijing, China). Fetal Bovine Serum (FBS, Catalog Number: A5256501) was purchased from Gibco (Waltham, MA, USA). DMEM High-Glucose Medium (Catalog Number: MA0212), Penicillin-Streptomycin Dual Antibody Sterile Solution (Catalog Number: MA0110), CCK-8 Kit (Catalog Number: MA0225), Live/Dead Cell Staining and Detection Kit (Catalog Number: MA0361) were purchased from Dalian Meilun Biotechnology Co., Ltd. (Dalian, China). NO ELISA Kit (Catalog Number: AD3124MO), TNF-α ELISA Kit (Catalog Number: AD3051MO), IL-10 ELISA Kit (Catalog Number: AD2837MO), and IL-1β ELISA Kit (Catalog Number: AD3364MO) were purchased from Andy Huatai Biotechnology Co., Ltd. (Beijing, China).

The RAW 264.7 cell line (Item Number: ml088508) was purchased from Shanghai Enzyme-Linked Biotechnology Co., Ltd. (Shanghai, China). Freund’s Complete Adjuvant (Item Number: F850325) was purchased from Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). Sodium Pentobarbital (Item Number: BC1040) was purchased from Beijing Puludun Biotechnology Co., Ltd. (Beijing, China). Diclofenac Sodium Cream was purchased from Suzhou Yifan Pharmaceutical Co., Ltd. (Hangzhou, China). The 4% Paraformaldehyde Fixative Solution was purchased from Fuzhou Xijing Biotechnology Co., Ltd. (Suzhou, China). IL-6 Immunohistochemistry Kit and TNF-α Immunohistochemistry Kit were purchased from Shanghai Jingfeng Biotechnology Co., Ltd. (Shanghai, China). The IL-1β Immunohistochemistry Kit was purchased from Shanghai Yaji Biotechnology Co., Ltd. (Shanghai, China).

Experimental Animals: Specific Pathogen-Free (SPF) male Sprague Dawley (SD) rats, weighing 180–220 g and aged 6–7 weeks, were purchased from Liaoning Changsheng Biotechnology Co., Ltd. (Production License: SCXK (Liao) 2020-0001, Benxi, China).

4.2. Material Preparation and Characterization

4.2.1. Synthesis and Characterization of HAMA

Weigh 6 g of Hyaluronic Acid (HA) was dissolved in 300 mL of purified water and stirred continuously at 4 °C for 16 h until complete dissolution was achieved. Subsequently, 201 mL of dimethylformamide (DMF) and 6.93 g of Methacrylic Anhydride (MA) were slowly added to form Solution A, which was reserved for subsequent use.

Weigh 6 g of NaOH was dissolved in 300 mL of purified water to prepare a 0.5 M NaOH solution. This alkaline solution was then added dropwise to Solution A under continuous stirring, and the pH was adjusted to maintain a range of 8–9. The reaction mixture was maintained at 4 °C with continuous stirring for 12 h. Following the reaction, HAMA was obtained by adding anhydrous ethanol to the polymer solution, followed by overnight incubation at 4 °C. The resulting precipitate was collected via centrifugation, washed three times with anhydrous ethanol to remove residual reagents, and subsequentlydialyzed against deionized water for 48 h using a dialysis membrane. Finally, the purified product was obtained by freeze-drying.

The freeze-dried HAMA was cut into small pieces, sputter-coated with gold, and examined using a field-emission scanning electron microscope (FE-SEM, JSM-7800F, JEOL Ltd., Imizu, Japan). Fourier-transform infrared (FTIR) spectroscopy (FTIR-650 spectrometer, Tianjin Gangdong Technology Co., Ltd., Tianjin, China) was used to characterize HA and HAMA, with spectra recorded over a scanning range of 400–4000 cm^−1^ at a resolution of 2 cm^−1^. For 1H Nuclear magnetic resonance (NMR) analysis (EFM-150MR NMR spectrometer, Echo, Tatsuno, Japan), both HA and freeze-dried HAMA were dissolved in deuterated water (D_2_O) at a concentration of 4 mg·mL^−1^, using tetramethylsilane as an internal reference standard. This analysis enabled the determination of the average number of hydroxyl groups (-OH) substituted by methacryloyl groups per disaccharide unit in the HA molecular chain. The degree of HAMA grafting was calculated according to Formula (1).

$$[eqn]$$

I_a_: The integration area of the methacryloyl methyl proton peak (-C(CH_3_) = CH_2_) in HAMA, I_h_: The integration area of the acetamido methyl proton peak (-NHCOCH_3_) in HA.

4.2.2. Synthesis and Characterization of GelMA

According to reference [42], 100 mL of phosphate-buffered saline (PBS) was placed in a 500 mL conical flask and preheated to 50 °C under constant temperature conditions. Then, 10 g of gelatin (Gel) was gradually added to the PBS solution with continuous stirring until complete dissolution was achieved. The flask was wrapped completely in aluminum foil to protect the reaction mixture from light exposure. Subsequently, 8 mL of methacrylic anhydride (MA) was added dropwise to the solution, and the reaction was carried out under stirring for 2 h. After the reaction, 100 mL of pre-warmed PBS (50 °C) was added to terminate the process. The resulting solution was dialyzed against deionized water for 5 days with twice-daily water changes (morning and evening), followed by freeze-drying to obtain the final GelMA product.

FTIR (FTIR-650 spectrometer, Tianjin, China) was used to analyze both Gel and GelMA. Spectra were recorded over a wavenumber range of 400–4000 cm^−1^ as a resolution of 2 cm^−1^. For ^1^H Nuclear Magnetic Resonance analysis (NMR), (EFM-150MR NMR spectrometer, Tatsuno, Japan) Gel and GelMA samples were dissolved in deuterated water (D_2_O) at a concentration of 4 mg·mL^−1^, with tetramethylsilane serving as the internal standard. The degree of substitution of methacryloyl groups on the ε-amino groups of lysine residues along the Gel backbone was evaluated, and the grafting ratio of GelMA was calculated according to Formula (2).

$$[eqn]$$

I_a_: The integration area of the methacryloyl methyl proton peak (-C(CH_3_) = CH_2_) in GelMA, I_r_: The integration area of the side-chain terminal -CH_2_- proton peaks from lysine and hydroxylysine residues in the Gel molecule.

4.2.3. Preparation and Characterization of MTX-NCs

MTX-NCs were prepared via the solvent–non-solvent precipitation method. Briefly, an appropriate amount of MTX raw material was dissolved in a small volume of NaHCO_3_ solution, followed by the addition of 5 mL of anhydrous ethanol. Separately, PVP K30 was dissolved in purified water under continuous stirring at a controlled temperature. The MTX-containing ethanol solution was then injected into the aqueous PVP K30 solution under increased stirring speed. The mixture was stirred for a predetermined duration to form MTX-NCs. The particle size, polydispersity index (PDI), and Zeta potential of the resulting MTX-NCs were measured using a Malvern dynamic light scattering instrument (nanoparticle size and zeta potential analyzer, Malvern, UK). Morphology and structural features were examined by transmission electron microscopy (Talos F200S TEM, Waltham, MA, USA). FTIR spectroscopy, DSC, and XRD were employed to characterize the MTX raw material, PVP K30, physical mixture of MTX and PVP K30, and the lyophilized MTX-NCs. Drug concentration was quantified by HPLC, and the drug-loading capacity (DLC) of MTX-NCs was calculated according to Formula (3).

$$[eqn]$$

The chromatographic conditions are as follows: the chromatographic column is a reverse-phase Eclipse XDB-C-18 column (250 mm × 4.6 mm; 5 μm), the mobile phase consists of 0.1% formic acid and acetonitrile, the detection wavelength is 305 nm, the flow rate is 0.8 mL·min^−1^, the column temperature is 35 °C, and the injection volume is 5 mL.

MTX-NCs Differential Scanning Calorimeter (DSC) Testing

The DSC instrument was utilized to conduct DSC measurements on MTX raw material, PVP K30, MNT, the physical mixture of MTX-NCs, and the lyophilized powder of MTX-NCs. The samples were placed in a crucible for processing and subjected to DSC scanning within the temperature range of 20~300 °C. The heating rate was set at 10 °C·min^−1^, and the N_2_ flow rate was maintained at 50 mL·min^−1^.

4.2.4. Fabrication and Characterization of Dual-Network Hydrogel Microneedles Loaded with MTX-NCs

The MTX-NCs were prepared as described in Section 4.2.3. An appropriate quantity of PVP K30 was mixed with the MTX-NCs and stored at 4 °C to allow for swelling. Subsequently, HAMA, GelMA, and the photoinitiator I2959 were added to the mixture. The solution was continuously stirred in a 35 °C water bath until homogeneous, followed by additional swelling at room temperature. The solution was poured into a microneedle mold and centrifuged at 3120 r·min^−1^ for 10 min to ensure complete filling of the PDMS mold cavities, particularly the needle tip regions. After ultraviolet (UV) irradiation to initiate crosslinking, a second portion of the same solution was added and centrifuged under identical conditions to fully fill the backing layer of the mold. Following another UV exposure step, DHMN@MTX-NCs were obtained. Blank dual-network hydrogel microneedles (DHMN) were fabricated using the same protocol without incorporation of MTX-NCs. The morphological structure of the HAMA/GelMA blend was observed using field emission scanning electron microscopy (FESEM). FTIR spectroscopy and DSC were conducted to characterize the MTX raw material, blank DHMN powder, a physical mixture of DHMN@MTX-NCs, and the final lyophilized DHMN@MTX-NCs. The DSC detection method is consistent with “Section MTX-NCs Differential Scanning Calorimeter (DSC) Testing”. Furthermore, the morphology, mechanical properties, drug release profile, stability, and swelling behavior of the DHMN@MTX-NCs were systematically investigated.

Investigation of Mechanical Properties of DHMN@MTX-NCs

Mechanical strength inspection

Using Parafilm^®^ film (Amcor, Zürich, Switzerland) as a skin simulant, the insertion characteristics of DHMN@MTX-NCs were investigated. Four Parafilm^®^ films were stacked together, and the microneedles were inserted into the Parafilm^®^ film in the same direction using the thumb. After maintaining this position for 30 s, the microneedles were removed, and the number of holes in each layer of the film and the number of layers pierced by the microneedles were recorded.

Performance evaluation of voltage transformers

The toughness of DHMN@MTX-NCs was evaluated through compressive property testing. The tip of DHMN@MTX-NCs was placed downward on a smooth surface, and a flat surface was added to the back of DHMN@MTX-NCs to ensure uniform stress. A certain weight of weight was placed, and the weight was gradually increased. After each 1–2 min of action, the weight was removed, and the change in the tip of DHMN@MTX-NCs was observed.

Stability Investigation of DHMN@MTX-NCs

Due to the influence of water content in HMN on the mechanical properties of microneedles, in order to investigate the stability of the moisture absorption capacity of DHMN@MTX-NCs during storage and transportation [43], HMN and DHMN@MTX-NCs were stored in environments with relative humidities of 33%, 65%, and 96% for 14 days, respectively (magnesium chloride was used to simulate an environment with 33% relative humidity, sodium nitrite to simulate an environment with 65% relative humidity, and potassium sulfate to simulate an environment with 96% relative humidity). The hydrogels were weighed every 48 h. The calculation of moisture absorption capacity is shown in Formula (4):

$$[eqn]$$

W_Initial_ and W_Final_ represent the initial and final weights of the hydrogel, respectively.

Determination of Swelling Rate of DHMN@MTX-NCs

The experimental operation for determining the swelling rate of microneedles was conducted according to reference [44]. HMN and DHMN@MTX-NCs were weighed in a dry state (M_0_), and then soaked in PBS solution at room temperature for 1, 2, 4, 8, 12, 18, 24, 32, 40, 50, 60, 90, 120, 150, and 180 min. After soaking, the microneedles were taken out and gently dried with filter paper to eliminate excess surface water. They were weighed again at each time point (M_t_), and the swelling percentage, namely the equilibrium swelling percentage, was calculated as shown in Formula (5).

$$[eqn]$$

4.2.5. Investigation on the Release Characteristics of Dual-Network Hydrogel Microneedles

Agarose hydrogel was selected to simulate skin [45] to study the drug release characteristics of HAMA hydrogel microneedles, GelMA hydrogel microneedles, and dual-network hydrogel microneedles (prepared by mixing the two). Rhodamine B was used to simulate the drug. Agarose powder was dissolved in boiled purified water, allowed to cool, and the agarose hydrogel was prepared. The hydrogel microneedles loaded with Rhodamine B were pressed onto the surface of the agarose hydrogel and observed continuously for 7 days. The agarose gel was cut open daily to observe changes in color intensity.

4.2.6. In Vitro Drug Release Study of Drug-Loaded Microneedles

An in vitro release of MTX and DHMN@MTX-NCs was evaluated using the dialysis method. A standard solution of MTX was prepared by accurately weighing 15 mg of MTX and dissolving it in 22 mL of pH 7.4 phosphate-buffered saline (PBS) containing 1% sodium dodecyl sulfate (SDS). This solution was transferred into a dialysis bag with a molecular weight cutoff of 14,000 Da, which was placed in 300 mL of release medium (pH 7.4 PBS containing 1% SDS) under constant agitation at 37 ± 0.5 °C. For the DHMN@MTX-NCs formulation, an equivalent amount corresponding to 15 mg of MTX was used. The microneedle patch was punctured through two layers of Parafilm^®^ membrane and positioned on the surface of 300 mL of release medium, with the needle tips touching the liquid surface, followed by incubation under the same conditions. The experiment was performed at an oscillation speed of 110 r·min^−1^ and maintained at 37 ± 0.5 °C. Samples of 5.0 mL were collected at predetermined time points: 0.08 (5 min), 0.25 (15 min), 0.5 (30 min), 1, 2, 4, 6, 8, 12, 24, 48, and 72 h. After each sampling, an equal volume of fresh release medium was replenished to maintain sink conditions. The samples were analyzed by HPLC to quantify drug content. The cumulative release amount (Q_n_, mg·mL^−1^) and cumulative release percentage (%) were calculated according to Formulas (6) and (7), respectively, to investigate their relationship with time (t, h).

$$[eqn]$$ $$[eqn]$$

Q_n_: The cumulative release amount for the i-th sampling point; C_n_: The MTX concentration in the release medium at each sampling point; C_i_: The i-th sample concentration; V_0_: The volume of the release medium; V_i_: Sample volume; Q_n_*: The cumulative release per unit area at the n-th time point; and Q_cast_: The initial drug dose.

The t−Q_n_* curve was plotted with time (t) as the horizontal coordinate and in vitro cumulative release degree (Q_n_*) as the vertical coordinate, and the curve was fitted to calculate the release equation.

4.2.7. In Vitro Transdermal Permeation Study of Drug-Loaded Microneedles

An in vitro transdermal permeation study was conducted for MTX, MTX-NCs, and DHMN@MTX-NCs using the Franz diffusion cell [46]. Excised skin was mounted between the donor and receptor compartments, with the stratum corneum oriented toward the donor compartment and the dermis in contact with the receptor compartment. The receptor chamber was filled with 15 mL of pH 7.4 PBS containing 1% SDS, ensuring complete contact with the skin without air entrapment. A volume of 3.0 mL of MTX solution, MTX-NCs suspension, or DHMN@MTX-NCs dispersion was applied to the donor compartment. The diffusion cells were placed in a water bath maintained at 37 ± 0.5 °C with continuous stirring at 260 r·min^−1^. Samples (3 mL) were withdrawn from the receptor compartment at predetermined time points: 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 24, 48, and 72 h. An equal volume of fresh receptor medium was added simultaneously. All samples were filtered through a 0.22 μm microporous membrane prior to HPLC analysis. The cumulative transdermal permeation amount Q_n_ (mg·cm^−2^) and the cumulative transdermal permeation percentage Q_n_* (%) were calculated according to Formulas (8) and (9), respectively. The relationship between Q_n_* (%) and time (t, h) was evaluated.

$$[eqn]$$ $$[eqn]$$

Q_n_: Cumulative transdermal amount per unit area at the nth time point; C_n_: Drug concentration measured at the nth time point; V: Added volume of Receiver solution; V_i_: Sampling volume per interval; A: Effective Transdermal Area: Q_n_: Cumulative transdermal permeation percentage per unit area at the nth time point; Q_cast_: Initial administered dose.

The t−Q_n_* curve was plotted with t as the horizontal coordinate and in vitro cumulative transdermal permeation (Q_n_*) as the vertical coordinate, and the curve was fitted to calculate the release equation.

4.3. Cell Experiments

4.3.1. Hemolysis Assay

A 2% red blood cell (RBC) suspension was prepared and separately mixed with blank dual-network hydrogel microneedles (DHMN) and the MTX-NCs-loaded dual-network hydrogel microneedles (DHMN@MTX-NCs). After ultraviolet irradiation for 1 h, the mixtures were serially diluted to final concentrations of 100, 75, 50, 25, 12.5, 6.25, and 3.125 mg·mL^−1^. The suspensions were gently vortexed to ensure homogeneity. A positive control group (0.5 mL of 2% RBC suspension + 0.5 mL deionized water) and a negative control group (0.5 mL of 2% RBC suspension + 0.5 mL normal saline) were established. All samples were incubated at 37 °C for 3 hto assess hemolysis, and photographs were taken after incubation. Subsequently, the samples were centrifuged, and the absorbance of the supernatant was measured at 540 nm. The hemolysis rate for each group was calculated according to Formula (10).

$$[eqn]$$

H_Sample_: The absorbance value of the experimental group sample.

H_Positive_: The absorbance value of the positive control group.

H_Negative_: The absorbance value of the negative control group.

4.3.2. RAW 264.7 Cell Culture and Polarization

RAW264.7 mouse macrophages (RAW264.7 cells), purchased from Shanghai Enzyme-Linked Biotechnology Co., Ltd. (Shanghai, China), as adherent cells, were cultured in high-glucose DMEM supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Cells were maintained in an incubator at 37 °C with 5% CO_2_, and used for subsequent experiments upon reaching appropriate confluence.

RAW 264.7 cells were polarized via LPS induction. A total of 5 × 10^5^ RAW 264.7 cells were seeded in 6-well plates and cultured overnight. When cell confluence reached approximately 70%, the culture medium was replaced with 2 mL of fresh complete medium containing varying concentrations of lipopolysaccharide (LPS). Experimental groups included: a normal control group, treatment groups exposed to LPS at 50 ng·mL^−1^, 100 ng·mL^−1^, 400 ng·mL^−1^, 700 ng·mL^−1^, 1 µg·mL^−1^, and 1.2 µg·mL^−1^. Each condition was performed in triplicate. After 24 h and 48 h of co-culture with LPS, changes in cellular morphology were observed under an inverted microscope. The supernatant from each well was then collected, and the nitric oxide (NO) content was quantified using a commercial assay kit to evaluate the efficiency of macrophage polarization.

4.3.3. Cytotoxicity Study

A total of 1 × 10^5^ RAW 264.7 cells were seeded into 96-well plates and cultured for 24 h. The original medium was removed and replaced with gradient-diluted sample solutions. The experiment was conducted on both non-polarized and LPS-polarized RAW 264.7 cells. The experimental groups included: blank group, control group, DHMN group, MTX group, HMN group, and combination group. Each group was set up in triplicate.

After 48 h of incubation, 10 μL of CCK-8 reagent was added to each well. Following an additional 1 h of incubation, the absorbance (A) was measured at 450 nm. Cell viability was calculated according to the following Formula (11).

$$[eqn]$$

A_Sample_: Solution Group; A_Blank_: Blank Group; A_Control_: Control Group.

4.3.4. Live/Dead Cell Assay

Polarized RAW 264.7 cells were seeded in a 12-well plate at a density of 2 × 10^5^ cells per well and cultured overnight. The cells were then divided into 16 experimental groups and treated with complete medium containing the following: a control group (PBS), HMN groups with MTX concentrations of 4.54, 9.09, and 13.63 μg·mL^−1^, and combination groups containing the same MTX concentrations (4.54, 9.09, and 13.63 μg·mL^−1^) supplemented with 0.74 μg·mL^−1^ ADA solution.

Each group was set up in triplicate. After 24 h of culture, 100 μL of the Calcein AM working solution was added to each well according to the manufacturer’s instructions. The plate was incubated in a humidified 37 °C, 5% CO_2_ incubator for 15–30 min. During this period, the cells were gently washed 2–3 times with PBS, with each wash step controlled within 1 min. Cell viability was assessed by confocal laser scanning microscopy (CLSM) using an excitation wavelength of 515 nm. Fluorescence images were captured and subjected to semi-quantitative analysis using Image J 1.54 software.

4.3.5. Cell Scratch Study

RAW 264.7 cells (5 × 10^5^) were seeded in 6-well plates and polarized with LPS to establish an inflammatory model. A sterile pipette tip was used to create uniform vertical scratches on the confluent monolayer. Loosely attached cells were removed by gentle washing with PBS. Fresh complete medium containing either PBS (control), DHMN@MTX-NCs at concentrations of 4.54, 9.09, or 13.63 μg·mL^−1^, or the combination formulation (MTX at 4.54, 9.09, 13.63 μg·mL^−1^ plus 0.74 μg·mL^−1^ ADA solution) was then added to the respective wells. Each treatment group was performed in triplicate. At 24 h post-treatment, cell migration into the scratched area was observed and photographed under an inverted microscope. Wound closure was analyzed semi-quantitatively using ImageJ software to measure the remaining gap area.

4.3.6. Anti-Inflammatory Effects on Cells

Polarized RAW 264.7 cells were seeded in 12-well plates at a density of 1 × 10^5^ cells per well. The experiment was initiated when the cells reached approximately 85% confluence. The experimental groups included: blank control group (untreated normal cells), the model group (LPS-polarized cells), HMN groups (treated with HMN containing MTX at 4.54, 9.09, and 13.63 µg·mL^−1^), and the combination groups (treated with the same MTX concentrations as the HMN groups plus an ADA solution at 0.74 µg·mL^−1^). Each group was set up in triplicate. After 48 h of incubation, the supernatant from each well was collected. The levels of released nitric oxide (NO) and the cytokines TNF-α, IL-10, and IL-1β were measured according to the respective assay kit protocols.

4.3.7. Western Blot

RAW 264.7 cells were seeded in 6-well plates at a density of 4 × 10^5^ cells per well and divided into five experimental groups. The blank Control group (Control): Untreated cells. Model Group (Model): Cells treated with LPS to establish an RA inflammatory cell model. Positive Control Group (Positive): Cells treated with LPS for 2 h, followed by intervention with a solution containing 7.31 mg·mL^−1^ of diclofenac sodium cream. DHMN@MTX-NCs Group: Cells treated with LPS for 2 h, followed by intervention with a solution containing 13.63 µg·mL^−1^ of the DHMN@MTX-NCs. Combination Group: Cells treated with LPS for 2 h, followed by intervention with a solution containing 13.63 µg·mL^−1^ of the DHMN@MTX-NCs and 0.74 µg·mL^−1^ of ADA solution.

Each condition was performed in triplicate. After 24 h, total proteins were extracted using an inducible iNOS and COX-2 protein extraction kit according to the manufacturer’s instructions. Protein concentrations were determined by BCA assay. Samples were denatured by heating at 100 °C for 10 min in a water bath, aliquoted, and stored at −80 °C until use. Proteins were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) at 120 V for 100 min under room temperature conditions. Separated proteins were transferred onto polyvinylidene fluoride (PVDF) membranes on ice at 300 mA for 1.5 h. Membranes were blocked with 5% skim milk in TBST for 1 h, washed briefly, and incubated with primary antibodies against iNOS and COX-2 (both diluted 1:1000) overnight at 4 °C. After five washes with TBST (5 min each), membranes were incubated with horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. Following another round of washing, immunoreactive bands were visualized using enhanced chemiluminescence reagent and detected with a gel imaging system. Band intensities were analyzed by ImageJ software for relative protein expression quantification.

4.4. Animal Experiments

4.4.1. Skin Insertion Capability of DHMN@MTX-NCs

Excised rat skin was obtained, and subcutaneous fat tissue was carefully removed. The skin samples were mounted with the stratum corneum facing upward. DHMN@MTX-NCs were placed vertically above the skin surface, and a standardized force was applied to insert the microneedles. After 1–2 min, the microneedle patch was removed, and the resulting micro-indentations on the skin surface were visually assessed under magnification.

The same microneedle insertion procedure was performed in vivo on live rats. Following application for 5 min, the microneedles patches were removed, and the skin’s healing response was monitored. Digital images of the puncture sites were acquired at 6 min intervals throughout the observation period until complete re-epithelialization was observed.

4.4.2. Skin Healing and Irritation Potential of DHMN@MTX-NCs

After insertion of DHMN@MTX-NCs into the rat skin, the application site was evaluated for signs of allergic or irritant reactions, such as erythema or edema, at 0 and 24 h.

4.4.3. Swelling Behavior of DHMN@MTX-NCs in Rat Skin

Following insertion into the dorsal skin of rats, the microneedle patch was secured with medical adhesive tape for 10 min and then removed. The microneedles were analyzed using field emission scanning electron microscopy (FESEM) to assess swelling or structural integrity of the needle array.

4.4.4. Histological Examination of Rat Skin Following DHMN@MTX-NCs Insertion

After microneedle insertion into the dorsal skin, rats were euthanized by cervical dislocation. The treated skin was excised, fixed in tissue fixative for 24 h, and subsequently embedded in paraffin. Tissue sections encompassing the microneedle insertion sites were cut using a microtome. For histological analysis, sections were stained with hematoxylin for 5 min, rinsed with water, differentiated in acid alcohol, blued in alkaline water, and rinsed again. After dehydration through an ethanol series, sections were counterstained with eosin for 5 min, cleared in xylene, mounted with a resinous mounting medium, and examined under a light microscope for imaging.

4.4.5. Establishment of the AA Rat Model

SPF-grade SD male rats, weighing 180220 g and aged 67 weeks, were housed in a laboratory environment with a temperature of 25 ± 0.5 °C, a relative humidity of 40%~60%, an alternating light-dark cycle, and free access to food and water. Prior to the experiment, the rats were acclimatized for one week.

The left plantar surface of the rat’s hind paw was disinfected with 75% ethanol. 0.1 mL of Freund’s Complete Adjuvant (FCA) was injected subcutaneously into the footpad to induce primary immunization. On day 8, a booster injection of 0.1 mL of FCA was administered at the same site. This procedure successfully established the Adjuvant-Induced Arthritis (AA) rat model.

4.4.6. Experimental Animal Grouping and Dosing Regimen

On day 15 after the second immunization, AA model rats were randomly divided into six groups (n = 6 per group): blank control group (non-modeled rats), model group, positive control group, DHMN@MTX-NCs group, ADA group, and combination group.

Based on the clinical treatment regimen for rheumatoid arthritis involving MTX and ADA, interventions were administered once weekly for a total of five cycles. The model group and blank control group received topical application of normal saline on the footpad. The positive control group was treated with 0.05 g of diclofenac sodium cream applied to the ankle joint. In the DHMN@MTX-NCs group, microneedles were inserted into the dorsal skin of the foot and secured with pressure-sensitive adhesive tape. The ADA group received a subcutaneous injection of 0.3 mL ADA solution (1.2 mg·mL^−1^) [47]. The combination group received both therapies: first, subcutaneous ADA injection, followed 4 h later by transdermal delivery via DHMN@MTX-NCs microneedle insertion, to prevent potential interference between administration routes.

4.4.7. Measurement of Rat Body Weight

Body weight was recorded once before primary immunization. From day 8 until the end of the experiment on day 49, body weight was measured every 4 days to monitor changes, serving as an indicator of successful model induction and therapeutic response.

4.4.8. Rat Toe Arthritis Score

According to the assessment guidelines from the European Alliance of Associations for Rheumatology and the American College of Rheumatology (EULAR/ACR) [48], arthritis scores were evaluated every 4 days starting from day 8 (after initial immunization). Each paw was scored from 0 to 4, yielding a maximum possible score of 16 per animal. The detailed scoring criteria are presented in Table 7.

4.4.9. Measurement of Rat Paw Volume

The volume of the left hind paw was measured using a plethysmometer based on the water displacement method. Starting on day 8 and repeated every 4 days, the left ankle joint was marked, extended, and immersed up to the same reference point in the apparatus to ensure consistent submersion depth. To minimize measurement variability, all assessments were performed by a single experimenter. The recorded volumes were used to evaluate the success of arthritis induction and the therapeutic response over time.

4.4.10. Measurement of Rat Paw Thickness

Paw thickness was measured using a vernier caliper to quantify edema. Measurements were taken at the mid-section of the left hind paw beginning on day 8 and repeated every 4 days. These data were used to monitor disease progression and treatment efficacy.

4.4.11. X-Ray Imaging

On day 49, rats were anesthetized with an intraperitoneal injection of 2% sodium pentobarbital. The left ankle joints were then examined using an X-ray imaging system to evaluate joint morphology and structural damage.

4.4.12. Analysis of Organ Indices

At the end of the treatment period on day 49, rats were fasted for 12 h and subsequently euthanized via intraperitoneal administration of an overdose of 2% sodium pentobarbital. The thymus and spleen were rapidly excised and weighed. The thymus index and spleen index for each rat were calculated according to Formulas (12) and (13), respectively.

$$[eqn]$$ $$[eqn]$$

4.4.13. Histopathological Analysis of Ankle Joint Tissue

Following euthanasia on day 49, intact ankle joints were dissected, and surrounding skin and soft tissue were removed. The joints were fixed in 4% paraformaldehyde, decalcified in 10% EDTA solution for several weeks, embedded in paraffin, and sectioned. Sections were stained with hematoxylin and eosin (H&E) and Safranin O/Fast Green to assess cartilage degradation, synovial inflammation, and proteoglycan loss for histopathological evaluation.

4.4.14. Ankle Joint Immunohistochemical Experiment

Ankle joint tissue sections were prepared as described in Section 4.4.13. After deparaffinization and rehydration, antigen retrieval was performed using EDTA buffer (pH 9.0) under microwave irradiation for 23 min. Endogenous peroxidase activity was blocked with 3% H_2_O_2_ in the dark for 25 min. Sections were then incubated with blocking solution containing 3% bovine serum albumin (BSA) in PBS for 30 min to reduce non-specific binding. Primary antibodies targeting IL-6, IL-1β, and TNF-α were applied and incubated overnight at 4 °C. After washing, sections were incubated with HRP-conjugated secondary antibody for 1 h at room temperature. Immunoreactivity was visualized using DAB substrate, yielding brown-yellow precipitates at positive sites. Sections were counterstained with hematoxylin for 3 min, dehydrated through graded alcohols, cleared in xylene, and mounted with neutral balsam. Images were captured using an Olympus light microscope (not fluorescence), and semi-quantitative analysis of staining intensity was performed using Image Pro Plus 6.0 software.

4.5. Statistical Analyses

All data were analyzed using SPSS version 23.0 software. Data are presented as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used for group comparisons. For homogeneous variances, post hoc pairwise comparisons were conducted using the least significant difference (LSD) test. The graphical abstract and animal images used in this study were obtained from the BioGDP website under appropriate licensing.

5. Conclusions

Based on the clinical treatment needs of RA, this study selected MTX as the model drug and combined microfabrication technology to construct a dual-network hydrogel microneedle system with high drug-loading efficiency and excellent delivery performance. Firstly, MTX-NCs with uniform particle size and stable dispersion were successfully prepared by the solvent–non-solvent precipitation method, with a drug-loading capacity of 61.3% (w/w). In vitro release results showed that the cumulative release of MTX-NCs within 72 h was 1.59 times that of free MTX, confirming that the nanocrystallization strategy significantly improved the poor solubility and slow release of MTX. Subsequently, MTX-NCs were efficiently loaded into hyaluronic acid/gelatin dual-network hydrogel microneedles by the centrifugal casting method. This dual-network structure endowed the microneedles with excellent mechanical strength, controllable swelling behavior, and good biocompatibility, which were significantly superior to those of traditional single-network hydrogel microneedles. In vitro cell experiments demonstrated that MTX-NCs-loaded dual-network microneedles could significantly inhibit the migration ability of RA-related inflammatory cells in a concentration-dependent manner. When combined with ADA, the synergistic effect was further enhanced, and the therapeutic effect was superior to that of the single-drug group. This microneedle system could bidirectionally regulate the inflammatory microenvironment: on the one hand, it upregulated the expression of the anti-inflammatory factor IL-10, and on the other hand, it significantly downregulated the expression of pro-inflammatory factors TNF-α, IL-1β, and downstream effector molecules NO, and inhibited the expression of key inflammatory pathway proteins iNOS and COX-2. In the AA mouse RA model, both the microneedle monotherapy group and the microneedle/ADA combination group could significantly alleviate joint swelling, inhibit bone erosion progression, and reduce synovial hyperplasia and cartilage destruction. Among them, the combination group had the most prominent therapeutic effect. In conclusion, the MTX-NCs dual-network hydrogel microneedle system constructed in this study not only achieved efficient transdermal delivery of MTX but also formed a dual intervention mode of local sustained release and systemic regulation through the targeted synergistic effect with ADA, providing a novel painless, precise, and efficient treatment strategy with clinical translation potential for RA.