A Programmed Drug‐Loaded and Penetration‐Delivery Functionalized Microneedle Patch for Synergistic Obesity Treatment

ABSTRACT

The inherent pathological microenvironment of adipose tissue restricts accumulation and deep penetration delivery of drugs, resulting in the lack of an ideal strategy for obesity management. In view of this, a dual‐layered microneedle system was developed for obesity treatment. Firstly, capsaicin (Cap) and mitochondrial uncoupler BAM15 were respectively co‐assembled with solutol HS‐15 (HS) to obtain functionalized nanodrugs (HS@Cap NPs or HS@BAM15 NPs). Then they were integrated into the microneedle system (BC‐MNs). The BC‐MNs realize programmed controlled release of nanodrugs in subcutaneous tissue via transdermal delivery. HS@Cap NPs rapidly release at the needle tip to penetrate deep white adipose tissue and promote its browning. Sequentially, slowly released HS@BAM15 NPs then accumulate in beige adipocytes to enhance cell metabolism, achieving synergistic anti‐obesity effects. Finally, both in vitro and in vivo experiments analyzed and validated the therapeutic efficacy of this system for obesity. In summary, the above‐mentioned cascade co‐treatment strategy based on the microneedle drug delivery system provides a potential possibility for clinical management and treatment of obesity.

Schematic diagram of double‐layered BC‐MNs with programmable sequential release for synergistic obesity therapy, where rapidly released HS@Cap NPs induces adipose browning and slowly released HS@BAM15 NPs boosts energy expenditure.

1. Introduction

Obesity, a chronic and recurrent metabolic disease caused by long‐term imbalance between energy intake and expenditure, has become a global health crisis [1, 2]. It is not only an independent disease state but also an important risk factor for type 2 diabetes, hypertension, non‐alcoholic fatty liver disease, cardiovascular diseases, and various cancers, imposing a heavy burden on the global public health system [3]. Conventional obesity management relies primarily on lifestyle interventions such as diet and exercise, which show low long‑term success and frequent weight regain under obesogenic environments [4, 5]. Most anti‑obesity drugs carry organ‑related side effects and fail to balance efficacy and safety [6, 7]. Although metabolic surgery is effective for severe obesity, it involves surgical risks, nutritional malabsorption, and complications, and is not suitable for all patients [8, 9]. Therefore, there is an urgent need to identify a safe and effective novel therapeutic strategy.

Current research indicates that adipose tissue, as a core energy metabolism organ, its dysfunction is the initiating link of obesity‐related metabolic diseases [10, 11, 12]. Moreover, adipose tissue is not an inert energy storage depot but an active endocrine and metabolic organ [13]. The discovery of beige adipose tissue (BAT) and the biological process by which white adipose tissue (WAT) can undergo browning to transform into beige fat (with thermogenic capacity) has opened up a brand‐new “fat‐burning” paradigm for obesity treatment [14, 15]. However, to transform above paradigm into clinically effective strategies, there are still some key scientific issues to be addressed. On the one hand, the WAT has a low degree of vascularization and relatively poor blood supply, making it difficult for browning drugs to reach the target sites sufficiently through the bloodstream [16, 17, 18]. Moreover, its heterogeneous tissue structure also poses a natural barrier to drug delivery [19, 20]. And traditional treatments mainly rely on systemic administration (such as oral or injection), which has a weak targeting effect on adipose tissue [21]. Only a small portion of the drugs can reach the adipose tissue at the lesion site, and their subsequent deep penetration ability is limited [22, 23]. Therefore, the actual weight loss effect is often unsatisfactory. Based on this, Zhen Gu and co‐workers encapsulated rosiglitazone nanoparticles (NPs) in microneedles (MNs) for sustained adipose browning with reduced systemic toxicity [24]. Additionally, several studies have demonstrated that loading browning agents such as capsaicin (Cap), metformin, β3‐adrenoceptor agonist, and Thyroid T3, Indocyanine green, and Au₂₅ Clusterzymes into MNs can undergo browning, offering promising implications for anti‐obesity therapies [25, 26, 27, 28, 29]. On the other hand, the poor water solubility and bioavailability of traditional inducers themselves result in relatively insufficient efficiency and persistence in inducing the browning of white fat, and the phenomenon of “whitening” reversal is prone to occur after the stimulation is terminated [30]. Given this, nanoscale encapsulation of browning agent enhances browning effects and improves anti‐obesity efficacy [31, 32, 33]. Briefly, several studies have constructed nano‐drug delivery systems loaded with browning agents (rosiglitazone, resveratrol) to promote adipocyte browning and upregulate UCP‐1 expression, thereby enhancing anti‐obesity effects [34, 35].

In addition, even with normal browning, lipid‐burning efficiency will decrease significantly without coordinated activation of mitochondrial lipid oxidation. Because UCP‐1 alone cannot effectively uncouple the mitochondrial respiratory chain and requires other uncoupling agents for synergistic enhancement [36, 37]. The classic mitochondrial uncoupling agent 2,4‐dinitrophenol has shown certain weight loss effects without changing food intake, but its narrow therapeutic window and high toxicity have limited further research on its application [38]. The latest research has found that N5, N6‐Bis (2‐Fluorophenyl)[1,2,5]oxadiazolo [3,4‐b] pyrazine‐5, 6‐diamine (BAM15), as an orally administered mitochondrial uncoupling agent, can reduce fat content in mice without affecting food intake, muscle mass, or body temperature, and has no obvious toxic side effects at high doses [39]. This weight loss method that increases BAT cell metabolism through mitochondrial uncoupling agents can burn fat without exercise, showing potential application value in current weight loss treatments [40]. However, BAM15's poor water solubility and short half‐life result in low bioavailability, which limits further research on its application. Overall, although certain achievements and breakthroughs have been made in strategies related to the browning of white fat or the fat‐burning of beige fat, there is still a lack of effective thinking and coordinated planning. As a result, the actual therapeutic effects remain limited.

Based on the above considerations, with the assistance of solutol HS‐15 (HS) as a midrange biologic drug transepithelial permeability enhancer, we developed functionalized double‐layered MNs system, which not only enables direct drug delivery to the adipose tissue through skin penetration but also allows for longitudinal drug delivery within it, thereby achieving synergistic treatment for obesity (Figure 1). Specifically, HS was first co‐assembled with Cap and BAM15 respectively to obtain nano‐drugs (HS@Cap NPs or HS@BAM15 NPs) with tissue‐penetrating properties. Then, HS@Cap NPs were dispersed in hyaluronic acid (HA) matrix to prepare the tip layer. HS@BAM15 NPs were mixed with hyaluronic acid methacryloyl (HAMA) to form the lower layer of the stratified microneedles, jointly constructing drug‐loaded functionalized double‐layered microneedles (BC‐MNs). After the BC‐MNs act on the adipose tissue, firstly, the HS@Cap NPs rapidly released from the HA needle tip layer could self‐permeate and deliver among the white adipose tissue and initiate the browning effect. Subsequently, the HS@BAM15 NPs slowly released from the lower layer of ultraviolet cross‐linked HAMA could also effectively penetrate the BAT and target the adipocytes to achieve the fat‐burning effect and stimulate energy consumption. In vitro and in vivo experiments also confirmed that the BC‐MNs could not only achieve programmed drug release but also promote the deep delivery of drugs among adipose tissues, thereby demonstrating a cascading and synergistic effect in obesity treatment. This provides a potential functional solution for the clinical treatment of obesity.

FIGURE 1.

Schematic diagram of the construction of BC‐MNs and its application in the synergistic treatment of obesity.

2. Results and Discussion

2.1. Fabrication and Characterization of MNs Loaded With NPs

By constructing a functionalized drug‐loaded MNs delivery system, it is possible to enhance the aggregation, penetration, and delivery of anti‐obesity drugs after local delivery to adipose tissue. In short, this treatment system combines HS with anti‐obesity drugs (BAM15 and Cap). HS is a nonionic amphiphilic surfactant, and its rich alkyl chains can assemble with hydrophobic drugs. While addressing the inherent problems of poor water solubility and low bioavailability of anti‐obesity drugs, it can achieve the penetration and delivery of drugs within adipose tissue [41, 42, 43]. The co‐assembly process of HS and anti‐obesity drugs is shown in Figure 2a. Transmission electron microscopy (TEM) confirmed that HS@BAM15 and HS@Cap NPs displayed similar to spherical (Figure 2b,c), and dynamic light scattering (DLS) indicated that both NPs possessed hydrodynamic diameters in the range of 14–15 nm (Figure 2d,e), confirming the excellent morphology and dispersion of the synthesized NPs. Subsequent UV–vis spectroscopy analysis further confirmed an absorption peak at 332 nm for HS@BAM15 NPs (Figure 2f; Figure S1a), primarily attributed to the BAM15 molecule. Meanwhile, HS@Cap NPs showed a characteristic absorption peak of Cap near 280 nm (Figure 2g; Figure S1b), both confirming the successful co‐assembly of the drugs with HS. Furthermore, the stability of the NPs in different physiological media was evaluated. Both HS@BAM15 and HS@Cap exhibited no significant changes in particle size after 24 h in ddH₂O, PBS, and PBS containing 10% FBS (Figure 2h,i), confirming the excellent stability of these NPs in physiologically relevant media.

FIGURE 2.

The fabrication and Characterization of MNs Loaded with NPs. (a) Schematic diagram of the co‐assembly of NPs. (b, c) TEM images of HS@BAM15 and HS@Cap NPs. (d, e) DLS of HS@BAM15 and HS@Cap NPs. (f, g) UV–vis of HS@BAM15 and HS@Cap NPs. (h, i) Particle size changes of HS@BAM15 and HS@Cap NPs in different physiologically relevant media. (j) Schematic diagram of BC‐MNs fabrication. (k) Optical images of BC‐MNs. (l) SEM images of BC‐MNs. Data are presented as mean ± SD, n = 3.

BC‐MNs were fabricated using HA and HAMA as matrix materials. The drug‐loaded HA layer releases the drug rapidly during degradation, while the cross‐linked HAMA layer slowly releases the encapsulated drug [44, 45, 46]. The preparation process of MNs is shown in Figure 2j. The BC‐MNs presented a uniform and complete 15 × 15 array (Figure 2k). The scanning electron microscope (SEM) images indicated that the MNs were conical, with a needle height of 1500 µm, a base width of 640 µm, and a center spacing of 920 µm (Figure 2l). More importantly, both HS@Cap and HS@BAM15 NPs exhibited high encapsulation efficiency (over 70%) in the NPs and MNs systems, laying a solid foundation for the subsequent transdermal delivery of the drugs and anti‐obesity therapy (Figure S2).

Additionally, we further verified the programmed drug release capability of the double‐layered MNs. We used FITC and 780 instead of the target drugs to prepare MNs, allowing us to visually track the in vitro drug release behavior. 780 was loaded in the HA layer, and FITC was loaded in HAMA. The results confirmed that 780 loaded in the HA layer was rapidly released in the buffer, while FITC in the UV‐crosslinked HAMA was released slowly (Figure S3). The in vitro release experiments demonstrated that the double‐layered MNs structure and the differences in matrix release kinetics enable programmed drug release, thereby achieving the goal of synergistic therapy.

2.2. Characterization and Implantation Efficacy of Double‐Layered MNs

BAM15 molecular structure contains element F, which is absent in Cap. Elemental analysis examined the distribution of the drug across different regions of the BC‐MNs. As shown in Figure 3a, the upper layer of the MNs tip constitutes Region 1, where elemental signals for C, N, and O were detected. The middle and lower layers near the MNs backing constitute regions 2 and 3, where C, N, O, and F signals were detected. with regions 2 and 3 containing the characteristic F element of BAM15. This confirmed that HS@BAM15 was primarily concentrated in the middle and lower layers of the MNs, whilst HS@Cap was mainly distributed in the upper layer near the MNs tip. To provide preliminary confirmation of the successful preparation of the layered BC‐MNs. Subsequently, visual MNs were prepared to further investigate the double‐layered structure, with NR and NB dyes used to replace Cap and BAM15, respectively. Three‐dimensional images of the MNs were observed via confocal laser scanning microscopy (CLSM), clearly confirming the bilayer structure of the MNs (Figure 3b). In summary, elemental analysis combined with visual imaging jointly verified the successful fabrication of the layered MNs.

FIGURE 3.

Characterization and implantation efficacy of double‐layered MNs. (a) Elemental analysis of BC‐MNs. (b) CLSM images of double‐layered MNs. (c, d) Mechanical properties and schematic diagram of BC‐MNs. (e) Percentage of BC‐MNS inserted into Parafilm M model. (f) SEM of MNs pressing into and out of porcine skin. (g) Fluorescence imaging of MNs implanted in mice. (h) CLSM images showed different depths of insertion into porcine skin. (i) Biocompatibility and skin safety of BC‐MNs after treatment. Data are presented as mean ± SD, n = 3.

MNs must possess sufficient mechanical strength to penetrate the skin's stratum corneum and deliver medication to the target site [47]. Force‐displacement test results showed that the mechanical strength of a single BC‐MNs needle reached 1.2 N, significantly higher than the critical insertion force required to penetrate the skin stratum corneum (Figure 3c,d) [48, 49]. Furthermore, Parafilm M, which has hydrophobic characteristics similar to the skin stratum corneum, is commonly used as a skin substitute model [50, 51]. Vertical compression tests were performed by applying pressure to BC‐MN onto layers of Parafilm M. The upper X‐axis represents the number of Parafilm M layers penetrated, and the lower X‐axis indicates the corresponding penetration depth. BC‐MN exhibited a penetration rate of 80% through eight layers of Parafilm M, with an estimated effective penetration depth of 1100 µm, demonstrating excellent mechanical properties (Figure 3e, Figure S4). When applied to fresh porcine skin, the corresponding results were presented in Figure S5. Meanwhile, SEM revealed morphological changes in MNs before and after insertion (Figure 3f). CLSM confirmed penetration to approximately 400 µm in porcine skin, supporting its use for obesity treatment (Figure 3h). The penetration depth achieved using porcine skin as a skin simulant was weaker than that of Parafilm M. This may be attributed to the elastic porcine skin providing resistance during the penetration process, while the high moisture content of fresh porcine skin may further influence the penetration depth.

To investigate drug distribution following different administration routes, MNs loaded with HS@780 were pressed into the inguinal adipose tissue of mice, while HS@780 NPs were administered via tail vein injection. Compared with the injection, MNs enhanced drug accumulation in inguinal adipose tissue, while no obvious fluorescence signals were observed in other major organs, thereby improving targeted drug retention and reducing systemic toxicity. Furthermore, we monitored the drug distribution at different time points after administration. At 4 h post‐administration, the fluorescence intensity in adipose tissue was significantly stronger in the MNs group than in the injection group, both groups showed decreased fluorescence signals at 16 h (Figure 3g; Figure S6). The safety of BC‐MNs administration was evaluated by observing the skin condition at the application site in mice. The result showed that skin punctures caused by MNs recovered almost completely within 60 min, with no signs of skin damage or inflammation in the treated area, confirming the excellent biocompatibility of the BC‐MNs system (Figure 3i).

2.3. Rapid Uptake and Intercellular Transfer of NPs

The biocompatibility of nano‐drug loaded composite MNs system was evaluated using CCK‐8 cell viability assays and hemolysis experiments. Results demonstrated excellent biocompatibility of the HS, NPs and MNs within the safety concentration range (Figure 4a–c; Figure S7). Subsequently, the uptake efficiency of BSA@780 and HS@780 NPs by 3T3‐L1 cells was measured at different time points after co‐culture. CLSM revealed that HS@780 was rapidly internalized by cells within 15 min, and the intracellular fluorescence intensity gradually increased with prolonged incubation. In contrast, no obvious fluorescence signal was observed in the BSA@780 group. These results confirmed that HS‐based NPs significantly promote the rapid intracellular uptake of drugs (Figure 4d,e).

FIGURE 4.

Rapid uptake and intercellular transfer of NPs. (a‐b) Relative cell viability of 3T3‐L1 cells co‐cultured with NPs and BC‐MNs (n = 3). (c) Hemolytic effects of NPs at different concentrations on red blood cells (n = 5). (d) Quantitative measurement of fluorescence intensity after uptake in 3T3‐L1 cells (n = 3). (e) Uptake of HS@780 NPs and BSA@780 NPs by 3T3‐L1 cells in different times. (f) Schematic diagram of cell transfer. (g) Quantitative measurement of fluorescence intensity transferred into 3T3‐L1 cells (n = 3). (h) CLSM images of intercellular transfer after treatment with HS@780 NPs and BSA@780 NPs. Data are presented as the mean ± SD. ****p < 0.0001 is considered statistically significant.

Research indicated that HS could disturb cell membrane structure to enhance membrane fluidity [52, 53]. To investigate the intercellular permeability of NPs after rapid uptake, the experimental workflow was designed as shown in Figure 4f. After incubating 3T3‐L1 cells with BSA@780, HS@780, and free 780 for 6 h, CLSM results indicated that 3T3‐L1 cells internalized the NPs and exhibited fluorescent signals (I). During the intracellular release phase (II) and the subsequent new cell reuptake phase (III, IV), the fluorescence signal observed at HS@780 remained persistently clear. In contrast, the fluorescence intensity of the BSA@780‐treated group and the free 780 treatment group was significantly weakened at the same time point. After 60 min of incubation, both BSA@780 and HS@780 exhibited fluorescence signals at 780, but the fluorescence intensity of HS@780 remained significantly stronger than that of BSA@780 (Figure 4g,h; Figure S8). These results indicated that the NPs prepared based on HS can be re‐released after taking up cells and transferred to surrounding cells, completing the reuptake and release cycle, providing a key experimental basis for obesity treatment strategies.

2.4. Anti‐Lipogenesis of NPs in Vitro

To evaluate the anti‐ lipogenesis effect of the NPs, the number and size of intracellular lipid droplets were determined using BODIPY fluorescence staining. Cell differentiation was performed following the MDI induction protocol, as illustrated in Figure 5a, which successfully induced the differentiation of preadipocytes (NC) into mature adipocytes (MA). The MA group exhibited significantly enhanced green fluorescence, indicating that NC cells were fully differentiated into MA. After treatment with Cap, BAM15, HS@Cap, and HS@BAM15 NPs, the fluorescence intensity and lipid droplet number in all groups were notably reduced. Notably, the NPs‐treated groups showed superior anti‐lipid droplet activity compared with the free drug groups (Figure 5b,d).

FIGURE 5.

Anti‐lipogenesis effects of NPs on MA under different treatments. (a) Schematic of MDI‐induced NC differentiation into MA. (b) Anti‐lipid droplet effects under different treatments by BODIPY staining (c) Immunofluorescence staining analysis of UCP‐1 expression. (d) Quantitative measurement of BODIPY‐stained fluorescence intensity. (e) Quantitative measurement of UCP‐1 fluorescence intensity. (f) Cellular ATP levels (%) after different treatments. (g) Expression levels of intracellular TG and CHO. Data are presented as mean ± SD, n = 3. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 are considered statistically significant, and ns, not significant.

Studies have demonstrated that cap promotes browning and suppresses lipid biosynthesis [54]. Here, we investigated the expression of the browning‐associated UCP‐1. As shown in the immunofluorescence results in Figure 5c, both Cap and HS@Cap enhanced the UCP‐1 fluorescence signal compared to the NC and MA treatment groups, with NPs exhibiting a stronger ability to promote adipocyte browning (Figure 5e). Previous studies have demonstrated that the mitochondrial uncoupler BAM15 reduces oxidative phosphorylation coupling efficiency, decreases ATP synthesis, and promotes energy expenditure, thereby inhibiting obesity [55, 56]. To further investigate the regulatory effects of BAM15 and HS@BAM15 NPs on adipocyte ATP levels, we co‐incubated them with MA for 12 h. Both BAM15 and HS@BAM15 significantly reduced ATP levels in adipocytes, with the HS@BAM15 exhibiting a more pronounced ATP downregulation effect than the BAM15 (Figure 5f).

Cholesterol (CHO) and triglyceride (TG) levels are key indicators for evaluating metabolic disorders. We further investigated the regulatory effects of different treatments of total CHO and TG in MA. The results showed that both Cap and BAM15 could reduce intracellular TG and CHO content, and the downregulating effects of HS@Cap and HS@BAM15 NPs were even more significant (Figure 5g). In summary, these results confirmed that HS co‐assembled drugs could efficiently preserve the bioactivity of Cap and BAM15, and their anti‐lipogenesis effects was significantly superior to the free drugs, achieving anti‐obesity treatment.

2.5. Tissue Penetration and Programmed Drug Release

The unique physiological microenvironment of adipose tissue forms a natural barrier to drug delivery. Consequently, efficient drug penetration and programmed release within adipose tissue constitute the core mechanism for anti‐obesity therapeutic efficacy. We developed a functionalized double‐layered MNs system that penetrates the skin barrier, directly delivers drugs to adipose tissue, and enables intratissue drug transfer (Figure 6a). Porcine adipose tissue was used to simulate the in vivo physiological environment, and the tissue penetration efficacy of HS@NR, BSA@NR, and free NR was evaluated via frozen sectioning. After 24 h of incubation, the fluorescence intensity and distribution range of HS@NR in adipose tissue were significantly superior to those of the other groups. These results were consistent with previous cell delivery experiments, demonstrating that HS‐assembled NPs can effectively improve the penetration depth of drugs in adipose tissue (Figure 6b; Figure S9).

FIGURE 6.

Tissue penetration and programmed drug release. (a) Schematic of tissue penetration and programmed drug release. (b) Penetration of HS@NR, BSA@NR, and NR in porcine adipose tissue. (c) Programmed drug release from double‐layered MNs in vivo.

To evaluate the programmed drug release behavior of double‐layered MNs, RN and RB were separately mixed with HAMA and HA to prepare visualized double‐layered MNs. After applying MNs to the inguinal adipose tissue of mice, HA rapidly released the encapsulated drugs by absorbing tissue fluid, while UV‐crosslinked HAMA hydrogel achieved sustained slow drug release through swelling [57, 58, 59]. The programmed delivery of drugs in tissue was realized via the layered structure and the differential release characteristics of the matrices. Based on this, drug release behavior at different time points was analyzed using frozen sections. Weak NR fluorescence signal was observed at 5 min after administration, while almost no NB signal was detected. At 30 min, NR fluorescence was significantly enhanced with weak NB fluorescence appearing, confirming the rapid release of HA‐loaded drugs and only a small amount of drug release from HAMA. At 90 min and 150 min after administration, the NB fluorescence signal continued to increase, further verifying the sustained release of HAMA‐loaded drugs in adipose tissue (Figure 6c). These findings collectively demonstrated that the double‐layered MNs drug delivery system achieves programmed drug release, improves drug penetration in adipose tissue, and exerts a cascading synergistic anti‐obesity effect.

2.6. The Effect of BC‐MNs on Weight Loss in Mice

We further evaluated the weight‐loss effects of different treatments in high‐fat diet (HFD)‐induced obese mice. Before the start of treatment, HFD‐induced obese mice were randomly divided into five groups (Figure 7a), including an untreated control group (Control), blank drug‐free HA‐HAMA MNs (Blank‐MNs), HS@BAM15‐loaded MNs (B‐MNs), HS@Cap‐loaded MNs (C‐MNs), and double‐layered MNs loaded with both HS@Cap and HS@BAM15 (BC‐MNs). Prior to administration, the inguinal region of the mice was shaved, and MNs were pressed onto the inguinal area of the mice (Figure 7b). After one month of treatment, body weight in Control and Blank‐MNs group recovered to 113.18% and 111.5% of baseline levels, respectively, while all other treatment groups exhibited weight loss. Among them, the BC‐MNs group showed the most significant weight loss, with body weight dropping to 84.02% of baseline levels (Figure 7c). Waist circumference measurements at the end of treatment revealed that all treatment groups exhibited significantly smaller abdominal waist circumferences compared to Control and Blank‐MNs groups (Figure 7d). Photographs of the mice taken on the final day of treatment further confirmed the weight‐reducing efficacy of the BC‐MNs (Figure 7e; Figure S10).

FIGURE 7.

Effects of different treatments on weight loss in HFD mice. (a) Schematic diagram of HFD mice model establishment and different treatments. (b) Photos of the administration site before and after MNs application. (c) Body weight changes in obese mice following one month of different treatments. (d) Waist size of mice after treatment. (e) Photos of mice after one month of treatment. (f) Food intake during the treatment period. (g) Total activity counts under a 24 h light‐dark cycle. (h) Oxygen consumption. (i) RQ. (j) Glucose tolerance test. (k) AUC of oxygen consumption. (l) AUC of RQ. (m) AUC of blood glucose level. Data are presented as mean ± SD, n = 3. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 are considered statistically significant, and ns, not significant.

At the end of the treatment period, metabolic cage assays were conducted on the mice. The results showed that the B‐MNs, C‐MNs, and BC‐MNs groups exhibited increased oxygen consumption (Figure 7h,k), decreased respiratory quotient (RQ) (Figure 7i,l), and food intake and total activity counts showed no significant differences across all groups (Figure 7f,g), indicating that the increased energy expenditure was not attributable to changes in food intake or activity levels. Among these, the BC‐MNs group exhibited the most significant effects, likely due to the synergistic action of the drugs on the layered MNs carrier, which achieved a stronger anti‐obesity effect by regulating energy metabolism.

In addition, the glucose clearance rate was improved in each treatment group after glucose injection. The effect was most obvious in BC‐MNs group. Compared with Control (18.1 mmol L⁻¹) and Blank‐MNs (18.3 mmol L⁻¹), the fasting blood glucose in the BC‐MNs group decreased to 12.3 mmol L⁻¹ (Figure 7j,m). The above results indicated that B‐MNs, C‐MNs, and BC‐MNs treatments could achieve anti‐obesity effects while effectively enhancing insulin sensitivity and inhibiting the further development of obesity‐related insulin resistance.

2.7. The Anti‐Fat Effect of BC‐MNs on Adipose Tissue

Following treatment completion, mice from each group were euthanized, and anatomy of four types of adipose tissue, including inguinal white adipose tissue (iWAT), perirenal white adipose tissue (pWAT), epididymal white adipose tissue (eWAT), and shoulder brown adipose tissue (sBAT). The volume and mass of the four adipose tissues in all treatment groups showed a downward trend, consistent with the changes in body weight and waist circumference. Among them, the reduction in the BC‐MNs group was the most effective (Figure 8a,b; Figure S11).

FIGURE 8.

Anti‐lipogenesis effect of different treatments on adipose tissue. (a) Adipose tissue was isolated and photographed after treatment. (b) Weight of adipose tissue. (c) H&E staining sections of adipose tissue. (d) Oil Red O staining images of sBAT. Data are presented as mean ± SD, n = 3. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 are considered statistically significant, and ns, not significant.

To further investigate morphological changes in adipose tissue, hematoxylin and eosin (H&E) staining was performed. Results demonstrated that all treatment groups significantly reduced adipocyte volume in the target site (iWAT), while also exhibiting marked reduction effects on adipocytes in pWAT, eWAT, and sBAT (Figure 8c). Comparing staining sections at 15 days post‐treatment with those at the end of treatment revealed further decreased adipocyte volume at the conclusion of therapy, confirming that the anti‐obesity effect of the MNs delivery system exhibits time‐dependent enhancement (Figure S12). Considering that sBAT is characterized morphologically by intracellular accumulation of multiple small lipid droplets [60, 61]. Further validation via Oil Red O staining revealed a reduction in adipocyte numbers within sBAT across all treatment groups, with the most pronounced decrease observed in the BC‐MNs group (Figure 8d). The above results all indicate that the microneedle drug delivery system treatment system not only achieves local targeting but also has the effect of regulating and reducing other adipose tissues.

2.8. Physiological Effects of BC‐MNs on Obese Mice

To verify the regulatory effects of each treatment on adipose tissue browning, UCP‑1 expression was examined by immunofluorescence. The results demonstrated that UCP‑1 expression was significantly upregulated in all treatment groups compared with Control and Blank‐MNs (Figure 9a). The locally delivered Cap within the therapeutic system promotes browning of WAT by activating dependent signaling pathways, thereby stimulating expression of the browning marker UCP‐1 [62, 63]. It is noteworthy that the mitochondrial uncoupler BAM15 also promotes UCP‐1 expression, thereby enhancing energy metabolism in mice [64]. UCP‐1 levels in BC‐MNs were higher than in the single‐component treatment groups (B‐MNs and C‐MNs). This may be attributed to the MNs in this system first releasing HS@Cap NPs to rapidly initiate fat browning, while the slowly released HS@BAM15 NPs continuously enhance the browning effect while accelerating energy expenditure, thereby achieving highly efficient and coordinated weight loss. Furthermore, we further evaluated the effect of different doses of BC‐MNs on UCP‐1 expression. The results showed that UCP‐1 expression was upregulated with increasing BC‐MNs dosage (Figure S13). H&E staining analysis also revealed that adipocyte size was reduced in a dose‑dependent manner (Figure S14). Collectively, H&E staining and UCP‐1 immunofluorescence jointly confirmed that both adipose browning capacity and therapeutic efficacy were gradually enhanced in a dose‑dependent manner.

FIGURE 9.

Physiological effects on obese mice after different treatments. (a) UCP‐1 staining of adipose tissue after different treatments. (b) Relative ATP levels in adipose tissue. (c) Changes in CHO, TG, LDL‐C, and GLU levels. (d) H&E staining of mice liver sections. Data are presented as mean ± SD, n = 3. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 are considered statistically significant, and ns, not significant.

Next, we examined ATP content in adipose tissue. Compared with the Control and Blank‐MNs, treatment groups were able to reduce ATP content in adipose tissue, with the BC‐MNs showing the best effect (Figure 9b). Administration of BAM15 disrupts the coupling between electron transport and ATP synthesis, thereby interfering with ATP production. Meanwhile, capsaicin stimulates ATP‐dependent thermogenesis by activating β3‐AR and TRPV1 channels, may lead to increased ATP consumption [65].

Finally, serum biochemical levels in mice were assessed. Compared with the Control and Blank‐MNs groups, CHO, TG, glucose (GLU), and low‐density lipoprotein cholesterol (LDL‐C) levels were reduced in the B‐MNs, C‐MNs, and BC‐MNs (Figure 9c). Concurrently, no abnormalities were observed in aspartate aminotransferase (AST), alanine aminotransferase (ALT), lactate dehydrogenase (LDH‐L), or alkaline phosphatase (ALP) levels across all treatment groups (Figure S15). Notably, the treatment groups effectively reduced serum AST and ALT levels, suggesting that the MNs drug delivery system may mitigate obesity‐associated liver injury. Subsequently, histopathological analysis of liver tissue revealed that the Control and Blank‐MNs exhibited numerous large vacuoles characteristic of typical fatty liver morphology. In contrast, all treatment groups showed a significant reduction in both the number and size of vacuoles (Figure 9d). This indicated that the drug‐loaded MNs system effectively suppresses abnormal hepatic lipid accumulation, thereby improving fatty liver development. Concurrently, histopathological examination of other vital organs revealed no significant abnormalities (Figure S16), confirming the system's favorable biosafety profile. H&E staining analysis of skin areas subjected to different treatments demonstrated intact skin tissue structure without evident damage such as inflammatory infiltration or tissue necrosis (Figure S17). This further validates the MNs’ excellent biocompatibility and indicates their substantial potential for obesity treatment.

3. Conclusion

We developed a double‐layered functionalized drug‐loaded MNs for cascading and synergistic treatment of obesity. Specifically, HS was assembled with Cap and BAM15 respectively to construct high‐adipose tissue permeable NPs (HS@Cap NPs and HS@BAM15 NPs), and they were integrated into the double‐layered MNs. Based on the hierarchical structure of MNs and the kinetics of matrix differential release, the programmed release of drugs has been achieved. BC‐MNs exhibited the following characteristics. First, the NPs co‐assembled with HS were rapidly internalized into cells and transferred between cells, demonstrating highly efficient penetration within adipose tissue. Second, in MDI‐induced adipocyte differentiation model, both HS@Cap and HS@BAM15 demonstrated significant anti‐adipogenic effects. Third, the programmed drug release within BC‐MNs enables rapid release of HS@Cap NPs from the tip layer to deeply penetrate iWAT and initiate browning. Concurrently, the sustained‐release HS@BAM15 NPs from the lower layer accumulate in beige adipose tissue, burning fat and accelerating energy expenditure to achieve synergistic, cascading weight‐loss effects. The constructed hierarchical microneedle drug delivery system offers a novel targeted delivery strategy for obesity treatment, demonstrating broad clinical application prospects.

4. Experimental Section

4.1. Materials

Solutol HS‐15 (HS), BAM15, and BODIPY 493/503 were purchased from MCE. Capsaicin (Cap), IR780 (780), Nile Red (NR) and Nile Blue (NB) were purchased from Sigma–Aldrich (USA). Fetal bovine serum (FBS) was purchased from Viacell (Shanghai, China). High‐glucose Dulbecco's Modified Eagle Medium (DMEM) was purchased from Biosharp (Beijing, China). Penicillin and streptomycin, trypsin, and Hoechst 33342 staining solution were purchased from Beyotime (Shanghai, China). 3‐Isobutyl‐1‐methylxanthine (IBMX), dexamethasone (Dex), bovine insulin, Fluorescein Isothiocyanate (FITC), and ATP content detection kits were purchased from Solarbio (Beijing, China). Hyaluronic acid Methacryloyl (HAMA, MW 150 kDa), Hyaluronic acid (HA, MW 150 kDa), and polyvinylpyrrolidone (PVA, MW 14 kDa) were purchased from Engineering for Life (Suzhou, China), Cholesterol (CHO) assay kit and triglyceride (TG) assay kit were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Ncoupled protein‐1 (UCP‐1) antibody was purchased from Santa Cruz (USA).

4.2. Synthesis and Characterization of Different NPs

4.2.1. Synthesis of HS@BAM15 and HS@Cap NPs

Weigh 120 mg of HS and place it into a centrifuge tube, then add 400 µL of anhydrous ethanol to dissolve it. Next, add 100 µL of BAM15 (DMSO, 5 mM) or Cap (DMSO, 5 mM) and mix thoroughly. Slowly add this mixture dropwise into 4 mL of ddH₂O, let it react vigorously for 6 h. After dialysis purification for 24 h, the NPs are successfully prepared. Store the NPs at 4°C.

4.2.2. Synthesis of other NPs (HS@780 or HS@NR or HS@NB NPs)

Replace the target drug used above with 100 µL 780 or 100 µL NR or 100 µL NB (DMSO, 10 mM) and prepare the NPs using the same method.

4.2.3. Synthesis of BSA@780 and BSA@NR NPs

Slowly add 100 µL of 780 or NR (DMSO, 10 mM) to 4 mL of bovine serum albumin (BSA, 5 mg mL⁻¹), then stir the mixture at room temperature for 12 h. Following the reaction, the mixture was preliminarily purified by dialysis in a dialysis bag for 24 h. The dialyzed solution was then transferred to an ultrafiltration tube (Millipore, 100 kd) for further purification, repeated 3 times. Finally, the prepared NPs were adjusted to a constant volume and stored at 4°C for subsequent use.

To characterize the morphology of NPs using TEM (FEI Tecnai, USA). DLS (Malvern, UK) was used to measure the particle size and dispersion. UV‐vis was utilized to detect their optical properties (Cary 5000, USA).

4.3. Fabrication and Characterization of the Double‐Layered MNs

HS@Cap NPs were dispersed in HA to prepare a 5% (w/v) polymer solution, which was then fully filled into a polydimethylsiloxane (PDMS) mold. After multiple vacuum degassing cycles, the mold was heated and concentrated in a vacuum drying oven to form the upper needle layer. Subsequently, HS@BAM15 NPs were dispersed in HAMA to prepare a 5% (w/v) polymer solution, together with photoinitiator (Irgacure 2959). Consistent with the methodology outlined above, UV curing was performed to form the lower layer of BC‐MNs. Finally, PVA solution was added as the backing layer. After drying and demolding, double‐layered MNs co‐loaded with HS@BAM15 NPs and HS@Cap NPs were prepared. Visualization of MNs was achieved by substituting NR and NB for the target drug using the same methodology.

The morphology and dimensions of the BC‐MNs were examined using SEM (Phenom prox, Netherlands). Elemental analysis of different regions of the BC‐MNs was performed via energy‐dispersive x‐ray spectroscopy (EDS) (Carl Zeiss, Germany). The layered structure of the MNs was observed and photographed using CLSM (Leica, Germany). The mechanical properties of the double‐layered were tested using an electronic universal testing machine.

4.4. Drug Loading Capacity (DL) and Encapsulation Efficiency (EE) of NPs and MNs

The DL and EE (in %) of Cap and BAM15 in NPs and MNs were further measured, with the corresponding calculation formulas presented below:

$$\mathrm{DL}=\frac{w_{\mathrm{encapsulated}}}{w_{\mathrm{NPs}\mathrm{or}\mathrm{MNs}}}\times 100\%$$

where w _encapsulated is the weight of Cap and BAM15 encapsulated in the NPs or MNs, and w _{NPs or MNs} is the total weight of the NPs or MNs.

$$\mathrm{EE}=\frac{C_{\mathrm{encapsulated}}}{C_{\mathrm{initial}}}\times 100\%$$

where C _encapsulated is the concentration of Cap and BAM15 encapsulated in the NPs and MNs, and C _initial is the initial concentration of the drug used for the preparation of NPs and MNs

4.5. In Vivo Fluorescence Imaging

To compare drug distribution between traditional injection and MNs delivery, HS@780 NPs‐loaded MNs were prepared and pressed onto the inguinal region of mice on one hand, while HS@780 NPs were intravenously injected via the tail vein on the other. Small animal in vivo imaging was used to observe the distribution of the dye at different time points. The mice's adipose tissue and various organs were dissected to observe the distribution of the drug in tissues and organs at different time points.

4.6. Skin Insertion Experiment

To investigate whether MNs can penetrate the skin for drug delivery. Double‐layered MNs loaded with HS@NR and HS@NB were prepared and pressed into porcine skin. After pressing, the MNs’ base layer was removed, and the penetration depth of the MNs was observed and photographed using CLSM.

4.7. In Vitro Drug Release

FITC and 780 were employed as model drugs to investigate the in vitro drug release from the constructed double‐layered MNs. BC‐MNs were placed in PBS containing hyaluronidase. At predetermined time points, 1 mL of PBS was aspirated and replaced with an equal volume of fresh PBS to maintain a constant system volume. The absorbance values of the drugs were measured using UV–vis. The cumulative release quantities of the model drugs were calculated to evaluate their in vitro drug release.

4.8. Adipose Tissue Penetration Assay

To compare the permeation capabilities of free drugs and different forms of NPs in adipose tissue, HS@NR, BSA@NR, and free NR were synthesized. They were added at equal concentrations to a transdermal delivery device containing adipose tissue. After 24 h of permeation treatment, the adipose tissue was removed, photographed, and examined via frozen sections to observe the permeation extent and depth.

4.9. In Vivo Programmatic Drug Release

To evaluate the programmed drug release capacity of drug‐loaded MNs, the as‐prepared MNs were applied to the inguinal region of mice. Inguinal adipose tissue was isolated at 5, 30, 90, and 150 min post‐administration to observe penetration intensity and depth.

4.10. Cell Culture

3T3‐L1 cells were seeded in high‐glucose DMEM supplemented with 10% FBS and 1% penicillin‐streptomycin solution, followed by routine culture in a humidified constant‐temperature incubator maintained at 37°C with 5% CO₂.

4.11. The Differentiation of Adipocytes

3T3‐L1 cells were cultured conventionally. Following 48 h of confluence, differentiation induction commenced by replacing the complete medium with a differentiation medium containing insulin (10 µg mL⁻¹), IBMX (0.5 mmol L⁻¹), and Dex (1 µmol L⁻¹). After 72 h of continuous culture in this high‐glucose differentiation medium, the medium was replaced with differentiation medium containing only insulin for a further 48 h. Finally, the cells were returned to complete medium until lipid droplets matured within the cells, signifying the differentiation of NC into MA.

4.12. Cytotoxicity Assay

3T3‐L1 cells were seeded into 96‐well plates and co‐incubated with NPs and MNs at various concentrations at 37°C for 24 h. Afterward, CCK‐8 reagent was added to each well, and following sufficient reaction, absorbance at 450 nm was measured using a microplate reader to calculate cell viability.

4.13. Rapid Cellular Uptake

To evaluate the uptake capacity of 3T3‐L1 cells for NPs, cells were cultured overnight in 35 mm confocal‐specific culture dishes. Subsequently, 0.5 µM concentrations of HS@780 NPs and BSA@780 NPs for 15, 30, and 60 min, respectively. Following incubation, cells were washed to remove residual drugs. Nuclei were specifically stained with Hoechst 33342, and drug distribution and cellular uptake were observed by CLSM.

4.14. Intercellular Transfer Experiment

To investigate the penetration ability of NPs in cells, 3T3‐L1 cells were seeded in confocal culture dishes and incubated overnight at 37°C. HS@IR780, BSA@IR780 and 780 were prepared at a concentration of 1.5 µM and incubated with the first batch of cells for 6 h. After PBS washing, fluorescence imaging was performed using CLSM. The cells were then cultured for an additional 12 h with fresh medium, after which the medium was collected, the cells were washed, and imaging was performed again. Meanwhile, the collected medium was added to a third batch of newly seeded cells, and after 12 h of continuous culture, cell imaging was carried out once more.

4.15. Lipid Profile Analysis

To validate the anti‐adipogenic effects of different treatments on adipocytes, lipid droplets were analyzed using the BODIPY assay after various treatments. NC were differentiated into MA via MDI induction. During differentiation, Cap, HS@Cap, BAM15, and HS@BAM15 were added and incubated. After treatment, the medium was discarded, cells were washed, and the BODIPY493/501 specific fluorescent probe was added for imaging. The fluorescence intensity after imaging was quantitatively analyzed. CHO and TG levels were measured post‐treatment using a kit.

4.16. UCP‐1 Cell Immunostaining Fluorescence

Cap and HS@Cap NPs were added to MA differentiated by the above method for incubation. After treatment, the cells were fixed with 4% paraformaldehyde, permeabilized at room temperature with 0.1% Triton X‐100 and blocked with 3% BSA. The treated cells were then incubated with a UCP‐1 antibody overnight at 4°C, followed by incubation with a fluorescent secondary antibody. Finally, the cells were mounted and imaged.

4.17. ATP Testing

After successfully differentiating NC into MA, add BAM15 or HS@BAM15 and incubate for 12 h. Subsequently, assess ATP levels in the adipocytes using an ATP assay kit.

4.18. Anti‐Obesity Experiments in Obese Mice Models

Male C57BL/6J mice (6 weeks old) were purchased from Chongqing Ensiwei'er Biotechnology Co., Ltd. and subjected to HFD feeding to induce obesity. All animal experiments were conducted under standard housing conditions, with animals maintained in appropriate environments and provided with free access to food and water. All procedures strictly adhered to the relevant regulations of the Animal Experiment Center Ethics Committee at Chongqing Medical University. The experimental protocol was approved by this committee (Approval No. IACUC‐CQMU‐2025‐0241).

The HFD mice were randomly divided into five groups, including the control group with Control, Blank‐MNs, B‐MNs, C‐MNs, and BC‐MNs, with 7 mice in each group. The drug‐loaded MNs were pressed onto the subcutaneous adipose layer of the inguinal region and administered bilaterally. Treatments were administered every two days for a period of one month. During the treatment period, the body weight, food intake, and waist circumference of the mice were continuously monitored following different treatments.

4.19. Animal Energy Metabolism Assessment

To clarify the differences in energy metabolism of mice after different treatments, indirect calorimetry was used to determine the metabolic parameters of mice. Specifically, three mice were randomly selected from each group, and their body weights were recorded. Each mouse was housed individually in a metabolic cage with stable airflow. After an appropriate acclimatization period to allow the mice to adapt to the housing environment, oxygen consumption, carbon dioxide production, RQ, spontaneous activity, and further analysis.

4.20. Glucose Tolerance Test

After fasting for 16 h, mice from different treatment groups were intraperitoneally injected with glucose (2 g kg⁻¹), and blood glucose levels in the tail vein were measured and recorded at specified time points.

4.21. Histopathological Assessment

After one month of treatment, blood was drawn from the eyeball and serum was separated for blood glucose and lipid testing, along with routine blood analysis. Subsequently, all mice were euthanized and dissected. Adipose tissue from various sites was isolated for immunohistochemical analysis and ATP content measurement, including inguinal iWAT, pWAT, eWAT, and sBAT. The collected adipose tissues and major organs were subjected to H&E staining. Additionally, skin tissue from the treatment area was collected to assess the material's biocompatibility.

4.22. Statistical Analysis

Unless otherwise specified, the results are presented as mean ± standard deviation (SD) with a minimum sample size of n ≥ 3 for each experiment. Comparisons between two groups were performed using Student's t‐test, and comparisons among multiple groups were analyzed by one‐way ANOVA. Statistical significance was indicated as follows: ns, not significant *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Statistical analyses were performed using GraphPad Prism and Origin.

Author Contributions

F.Z., Y.Y.Z., and Q.L. designed the experiments. F.Z. and Q.L. conducted the experiments, analyzed the data. Y.Q.X., H.H. and J.Y.Z. assisted in conducting the experiment. F.Z. wrote the manuscript. Y.Y.Z. and R.T. provided experimental materials. F.Z. provided analytical methodologies. Y.Y.Z., R.T., and H.H. supervised the project. Y.Y.Z. and Q.L. critically revised the manuscript. All authors reviewed and approved the final version of the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adhm71409‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.