Mesoporous polydopamine and inulin hydrogel for improved deoxynojirimycin effect in type 2 diabetes mellitus management

Abstract

Effective oral delivery of natural products is essential for sustained therapeutic effects for type 2 diabetes mellitus (T2DM) therapy. However, compromised bioavailability leads to undesirable effect and hypoglycemia. Herein, biomimetic nanoplatform based on mesoporous polydopamine (MPDA) and inulin gel were designed to enhance the oral delivery of 1-Deoxynojirimycin (DNJ). DNJ was adsorbed via mesopores and hydrogen bonding by MPDA and encapsulated within the inulin hydrogel. The MPDA@Inulin gel exhibited sustained-release properties that allowed DNJ to be released persistently in the small intestine, thereby enhancing its oral bioavailability. Moreover, the capacity of DNJ@MPDA@Inulin gel in controlling the blood glucose levels was evaluated in a high-fat diet induced T2DM model. Notably, after 4 weeks of daily oral administration, DNJ@MPDA@Inulin demonstrated enhanced efficacy in controlling blood glucose levels in T2DM model mice, improving insulin sensitivity. More important, this hydrogel nano-sustained release platform based DNJ delivery system could ameliorate glucolipid metabolism, enhance intestinal mucosal barrier and reshape gut microbial composition. Thus, it is anticipated that inulin-MPDA based oral delivery system have a potential for natural products delivery to improve bioavailability and alleviate T2DM.

Graphical abstract

1. Introduction

Diabetes mellitus (DM) is a progressive, complex metabolic disorder characterized by insulin deficiency or resistance [1]. As reported in the International Diabetes Federation's 2025 IDF Diabetes Atlas, type 2 diabetes mellitus (T2DM) accounts for over 90% of global diabetes cases, which associates with various life-threatening secondary health conditions [2]. Current treatment strategies primarily focus on direct oral administration of hypoglycemic drugs, such as sodium-glucose cotransporter-2 (SGLT2) inhibitors [3] and glucagon-like peptide-1 receptor (GLP-1R) [4] agonists to control blood sugar. However, due to the low bioavailability, it is difficult to achieve desirable effect and may cause severe hypoglycemic events for long term administration. Consequently, natural products as safe alternatives provide benefits to prevent or alleviate T2DM-related complications.

1-Deoxynojirimycin (DNJ) is a representative iminosugar, which naturally exists as a secondary metabolite in plants, insects, and microbial culture broths. As a glucose analog, DNJ is well-known for its potent α-glucosidase inhibitory activity and broad applications in control postprandial blood glucose [5], oxidative stress [6], and inflammation [7]. Compared to conventional hypoglycemic agents such as metformin, acarbose and insulin, DNJ exhibited enhanced potential in alleviating both diabetic liver and kidney injuries [7,8], which exhibiting a lower risk of inducing hypoglycemia [9]. Despite potent α-glucosidase inhibition in vitro, DNJ exhibited limited in vivo efficacy due to its short half-life [10,11]. Following oral administration, DNJ could achieve maximum plasma concentration within 30 min before undergoing rapid renal elimination [12,13]. The short biological half-life and oral pharmaceutical formulation difficulties was due to simple structure and high hydrophilicity [14,15], result in poor bioavailability and necessitating frequent high-dose administration. These limitations underscore the critical need for advanced oral delivery systems to optimize the therapeutic potential of DNJ. Therefore, to prolong the intestinal retention of DNJ and enhance its oral bioavailability is an innovative strategy for improving therapeutic efficacy for T2DM.

To address the challenges associated with the delivery of highly hydrophilic natural products, several advanced delivery systems, including lipid-based colloidal systems [16], porous nanocarriers [17], and hydrogel-based embedding platforms [18] were developed and evaluated. Among various nano-delivery systems, mesoporous polydopamine (MPDA) nanocarriers stand out for its excellent biocompatibility and strong drug loading ability [19]. However, orally administered nanocarriers exhibit limited gastrointestinal retention, requiring frequent dosing to sustain therapeutically effective concentrations.

Polysaccharide hydrogels, such as alginate, chitosan and hyaluronic acid have emerged as promising oral drug delivery vehicles for their capability to protect drugs from degradation by gastric acid and enzymes [20,21]. Inulin is a dietary polysaccharide derived from chicory and Jerusalem artichoke, which can self-assemble into a hydrogel and exhibit pronounced resistance to the harsh gastrointestinal environment and prolonged intestinal retention [22,23]. Furthermore, inulin can be degraded by colon-specific inulinase and inulin metabolites have recognized for its prebiotic effects, which selectively promote the growth of beneficial bacteria like Bifidobacterium [24]. Accumulating evidence revealed that the gut microbiota played critical role in T2DM development through the gut-liver axis, where specific microbial species and their metabolites could influence glucose homeostasis through metabolic pathways [25,26]. Thus, inulin-based gel systems for the co-delivery of natural active compounds, coupled with their inherent capacity to restore gut microbiota balance, presents a feasible and promising therapeutic strategy for the management of diabetes.

Herein, mesoporous polydopamine nanospheres (MPDA) was applied as a carrier to load DNJ (DNJ@MPDA), combined with inulin hydrogel (DNJ@MPDA@Inulin gel) for oral DNJ delivery (Scheme 1A). Furthermore, as described in Scheme 1B, the hydrogel nanoplatform could effectively protect DNJ in gastric acid environments, and obtain sustained-release of DNJ at small intestinal. A single oral administration of DNJ@MPDA@Inulin gel could significantly alleviate post-meal blood glucose levels in short-term hyperglycemic model. The chronic oral administration of the DNJ@MPDA@Inulin gel was also shown to maintained blood glucose levels, improved insulin sensitivity, regulated glucolipid metabolism, repaired intestinal mucosal barrier function, and modulated gut microbiota in T2DM mice. This work provides a general strategy based on MPDA@Inulin gel sustained-release delivery system to collaboratively contribute in the bioavailability of DNJ and treatment of T2DM remission.

Scheme 1.

MPDA@Inulin gel sustained-release system for DNJ delivery in T2DM alleviation. (A)Schematic Representation of DNJ@MPDA@Inulin gel Prepared (B) DNJ@MPDA@Inulin gel Alleviates T2DM by Suppressing α-Glucosidase, Remodeling the Glucolipid Modulation and Regulating the Intestinal Flora and Recovery Pancreas and Kidney dysfunction.

2. Results and discussion

2.1. Fabrication and characterization of MPDA and DNJ@MPDA nanoparticles

Mesoporous polydopamine (MPDA) serves as a multifunctional organic biopolymer formed through the self-polymerization of dopamine under alkaline conditions [27]. Based on large surface area from mesopores with dopamine-derived excellent biocompatibility and versatile chemistry, MPDA potentially offers capacity for drug retention and release, supporting prolonged drug delivery [28]. Therefore, MPDA was designed to encapsulate DNJ (DNJ@MPDA) for achieving sustained DNJ delivery. Initially, we explored the fabrication of MPDA nanoparticles by the traditional soft-template method as shown in Fig. 1A. The morphological analysis of MPDA via transmission electron microscopy (TEM) and scanning electron microscopy (SEM) showed that MPDA exhibited uniform spherical shapes with a mesoporous structure and an average diameter of approximately 180 nm (Fig. 1B and C). The dynamic light scattering (DLS) measurement revealed that the particle sizes of MPDA were 205.58 ± 6.25 nm (Fig. 1D), and the particle sizes were consistent with TEM and SEM results. UV-Vis spectroscopy revealed concentration-dependent absorbance of MPDA, exhibiting a prominent absorption peak at 290 nm where a robust linear correlation between absorption intensity and concentration was established (Fig. S1).

Fig. 1.

Synthesis and characterization of MPDA and DNJ@MPDA. (A) Synthetic route for mesoporous polydopamine nanospheres (MPDA) based on the versatile nanoemulsion assembly approach; (B) Images obtained via transmission electron microscopy (TEM), (C) scanning electron microscopy (SEM), and (D) particle size distribution determined by dynamic light scattering (DLS); (E) SEM image of DNJ@MPDA; (F)The particle size and (G)zeta potential changes of MPDA before and after DNJ adsorption; (H) α-Glucosidase inhibitory activity of residual DNJ in the supernatant after MPDA encapsulation; (I) Encapsulation efficiency of different MPDA: DNJ mass ratios; Data are expressed as mean ± SD (n = 3 biological replicates for F-I) and analyzed by one-way ANOVA. ***p < 0.001.

Subsequently, DNJ was loaded onto MPDA through mesopore adsorption and hydrogen bonding, and obtained the DNJ@MPDA. Briefly, DNJ and MPDA were dispersed in UP water under magnetic stirring to form DNJ@MPDA nanoparticles, where MPDA's phenolic hydroxyl groups and mesoporous structure facilitated both DNJ adsorption (Fig. 1A). The adsorption of DNJ did not significantly alter the surface morphology (Fig. 1E), particle size (Fig. 1F), or zeta potential (Fig. 1G) of DNJ@MPDA, demonstrating that DNJ loading preserves the structural stability of MPDA nanoparticles. After centrifugation of DNJ-loaded MPDA, using the optimized α-glucosidase inhibition rate reaction system, the supernatant and original DNJ solution were compared for α-glucosidase inhibitory activity (Fig. S2). As shown in Fig. 1H, α-Glucosidase inhibitory activity of residual DNJ in the supernatant after MPDA encapsulation decreased significantly from 68.84% to 48.61%, confirming successful DNJ adsorption onto MPDA. Furthermore, by continuously increasing the MPDA/DNJ mass ratio from 0.25: 5 to 1: 5, the DNJ encapsulation rate decreased from 15.92% to 7.28%. The encapsulation efficiency of DNJ by MPDA decreases as the amount of MPDA increases, possibly due to the aggregation of excess MPDA, which reduces the effective binding sites for DNJ (Fig. 1I). Additionally, the maximum encapsulation rate of MPDA was only 15%, indicating that the highly hydrophilic nature of DNJ makes it prone to detachment from MPDA. Notably, none of the formulations achieved a DNJ loading efficiency exceeding 20%, necessitating alternative strategies to improve DNJ payload.

2.2. Fabrication and characterization of DNJ@MPD@Inulin gel

Inspired by the excellent hydrogel forming and probiotic modulation capacities of inulin, inulin was further chosen to embed nano-adsorbed drug, thereby improving the loading content of DNJ and enhancing the small intestinal retention time. Firstly, thermally induced method was applied to prepare inulin hydrogels [29,30]. Briefly, inulin molecular chains extended and became fully hydrated during heating (≥70 °C); upon cooling, three-dimensional porous structures were established through hydrogen bonding [31]. Simply, the obtained DNJ@MPDA nanoparticles was incorporated into a preheated and cooled to room temperature inulin solution, forming the DNJ@MPDA@Inulin gel, followed by immediate homogenization and 24 h storage at 4 °C (Fig. 2A). The gel presented a pale grayish-yellow appearance (Fig. 2B), resulting from the presence of DNJ@MPDA. Fourier transform infrared spectroscopy (FTIR) assay and zeta potential measurements verified the successful preparation of the DNJ@MPDA@Inulin gel (Fig. S3A and B).

Fig. 2.

Synthesis and characterization of DNJ@MPDA@Inulin gel. (A) Inulin gel was prepared via conventional method, and liquid DNJ@MPDA was added and mixed before cooling to gel; (B) Two different inulin gel formulations (37.5 wt% inulin gel and 37.5 wt% inulin gel embedding DNJ@MPDA) were characterized by visual inspection, injectability images, and SEM analysis. (C, D) Dynamic rheological tests were performed on these two groups and their corresponding 27.5 wt% inulin hydrogels. (C) Flow sweep; (D) Frequency sweep of the hydrogels (G′ indicates the storage modulus, and G’’ indicates the loss modulus). (E) SEM images of 37.5 wt% inulin gel and 37.5 wt% inulin gel embedding DNJ@MPDA (DNJ@MPDA@Inulin gel) after treatment in SGF and SIF solutions for 2 and 6 h. (F) In vitro degradation of the DNJ@MPDA@Inulin gel was analyzed across a range of pH values.(G) DNJ release behavior was evaluated under various pH conditions at 37 °C. The data are presented as the mean ± SD (n = 3 biologically independent samples for (F, G)).

The hydrogel formulations exhibited characteristic porous hydrogel architectures, with negligible morphological variations irrespective of DNJ or MPDA incorporation as evidenced by the SEM images (Fig. 2B and Fig. S4A). Impressively, the DNJ@MPDA@Inulin gel exhibited excellent injectability, demonstrating significant potential for oral delivery applications. Furthermore, as indicated in Fig. 2C the storage moduli (G′) of inulin hydrogel was higher than loss moduli (G″), which suggested that DNJ@MPDA incorporation did not affected gelation properties. The conserved shear-thinning behavior further evidenced mechanical stability (Fig. 2D). Compared to the 37.5 wt% inulin gel, the 27.5 wt% inulin exhibited significantly lower mechanical strength and stiffness, as evidenced by its reduced storage modulus (G′), loss modulus (G″), and steady shear viscosity, indicating a looser three-dimensional network structure and weaker rheological properties. Therefore, increasing the inulin proportion in the inulin gel can enhance its mechanical strength.

The extreme pH environment in the gastrointestinal tract poses significant challenges for oral drug delivery. To evaluate the stability of the DNJ@MPDA@Inulin gel, simulated gastric fluid (SGF), simulated intestinal fluid (SIF), and neutral PBS buffer (pH 7.4) were applied as test solutions. Notably, SEM observations revealed that after 6 h of immersion in different physiological solutions, the 27.5 wt% Inulin gel exhibited significantly enlarged pore structures in SGF, indicating structural instability (Fig. S4B). However, when the Inulin concentration was increased to 37.5 wt%, both the DNJ@MPDA@37.5 wt% Inulin gel and the pure 37.5 wt% Inulin gel maintained intact microstructures (Fig. 2E), and quantitative porosity analysis further corroborated these findings. For the 27.5 wt% inulin gel, the porosity in SGF was significantly higher than that in PBS, reflecting the substantial impact of the surrounding medium on the weak network structure (Fig. S4C). In contrast, both the 37.5 wt% pure inulin gel and the DNJ@MPDA@Inulin gel (37.5 wt%) exhibited comparable porosity levels across PBS, SGF, and SIF, with no statistically significant differences, demonstrating that the higher inulin concentration confers enhanced structural stability and resistance to swelling/degradation in diverse physiological environments (Fig. 2F). Furthermore, FTIR analysis confirmed that the composition of the gel was unaltered under these conditions, as depicted in Fig. S5. The possible reasons may attribute to inulin's fructose units linked by β-(2 → 1) glycosidic bonds, which makes it resistant to hydrolysis by digestive enzymes in the gastrointestinal tract. Collectively, based on the rheological analysis and the stability assessments in SGF, the 37.5 wt% inulin formulation was selected for subsequent experiments due to its superior mechanical strength and structural integrity. Degradation studies further confirmed that the inulin gel was more susceptible to structural disruption in the acidic SGF environment (pH 1.5) compared to other media, this might be related to the acid sensitivity of the β-glycosidic bonds of inulin. Nevertheless, the degradation rate remained below 5% after 12 h and under 15% after 18 h (Fig. 2F), indicating that the controlled degradation profile enables effective protection of DNJ@MPDA during the gastric transit period, ensuring intact delivery to the intestinal tract. Furthermore, the DNJ release behavior in MPDA@Inulin was monitored. Existing studies report that free DNJ, once orally administered to mice, declines rapidly in the small intestine within 45 min [6]. To synchronize drug exposure with post-prandial glucose excursions, we engineered the MPDA@Inulin hydrogel to release approximately 70 % of its DNJ payload within 6 h (Fig. 2G), a timeframe that sharply contrasts with the complete release observed for the free compound within 2-3 h. Notably, as verified in Fig. S6, the DNJ@MPDA@Inulin gel exhibited significantly slower drug release than the pure DNJ@Inulin gel at 2 h under all simulated pH conditions, confirming the synergistic effect of MPDA and the inulin gel matrix in enhancing the sustained-release performance. This design sustains therapeutic concentrations over a full inter-meal interval, thereby validating the platform as a robust carrier for meal-matched, prolonged DNJ delivery. Collectively, to address the oral delivery of DNJ with simple structure and high hydrophilicity, MPDA based nano-platform and inulin hydrogel were successfully constructed, which demonstrated superior stability and sustained-release across simulated physiological environments.

Previous studies indicated that gel-state polysaccharide fibers could slow gastric emptying by increasing viscosity, adhere to the intestinal mucosa, and extend the duration of intestinal retention [22,23]. We further evaluated the small-intestine-retentive properties of Cy7-MPDA@Inulin gel in vivo (Fig. 3A). The distribution of Cy7-MPDA nanoparticles and Cy7-MPDA@Inulin gel in the digestive tract was monitored in real time in mice using an in vivo imaging system (IVIS). In vivo fluorescence imaging showed that oral delivery of Cy7-MPDA@Inulin gel substantially increased the residence time of the inulin gel throughout the small intestine (Fig. 3B). In particular, the Cy7-MPDA@Inulin gel could significantly improve retention between 2 and 4 h compared to Cy7-MPDA nanoparticles (Fig. 3C). At 6 h, the Cy7-MPDA@Inulin gel and Cy7-MPDA nanoparticles nearly disappeared strong fluorescent signals in the Gastro-Intestinal Tract (GIT). Collectively, the results demonstrated that the DNJ@MPDA@Inulin gel facilitated sustained DNJ retention and release in the small intestine, thereby enhancing systemic DNJ bioavailability.

Fig. 3.

Hydrogel increases the residence time in the small intestine. (A) Mice were orally administered with Cy7-MPDA nanoparticles and Cy7-MPDA@Inulin gel. (B, C) The gastrointestinal tract was monitored over a 6-h period (B), and the mean fluorescence intensity in the colon was measured (C). Data are expressed as mean ± SD (n = 3). Representative images from one of three independent experiments are shown. Statistical significance was determined using two-tailed Student's t-tests.

2.3. The DNJ@MPDA@Inulin gel significantly enhanced the sustained AGI activity and bioavailability of DNJ

α-Glucosidase inhibitors (AGIs) competitively and reversibly inhibit intestinal α-glucosidases, thereby delaying the breakdown of polysaccharides and disaccharides into glucose [32,33]. By suppressing these enzymes, AGIs attenuate postprandial glucose (PPG) excursions and improve glycemic control. DNJ is a well-characterized AGI that mitigates PPG fluctuations [34]. To evaluate the PPG-modulating efficacy of DNJ@MPDA@Inulin gel, we established a short-term hyperglycemic model via oral sucrose administration (Fig. 4A). According to the previous study that the half-life of DNJ was approximately 3 h [11],we administered a second sucrose challenge at the 3-h mark to maintaining a high blood sugar level and assess sustained efficacy. Fig. 4 B, C demonstrated that while DNJ@MPDA or DNJ@Inulin gel alone showed comparable PPG attenuation to free DNJ, the DNJ@MPDA@Inulin gel significantly outperformed free DNJ in suppressing glycemic excursions after the second sucrose challenge. To corroborate this in vivo efficacy, pharmacokinetic analysis was performed in mice (Fig. S7). The results revealed that free DNJ was rapidly cleared from circulation, whereas the DNJ@MPDA@Inulin gel exhibited a rebound in blood concentration at 4 and 6 h, indicative of colon-targeted drug release and the sustained-release effect of the MPDA@Inulin gel. These findings validated that the hybrid DNJ@MPDA@Inulin gel could improve the duration and potency of DNJ's AGI activity in vivo through sustained release, ultimately improving systemic DNJ bioavailability.

Fig. 4.

Effect of DNJ@MPDA@Inulin gel of the glucose lowering effects in vivo. (A) Roadmap for acute hyperglycemia experimental design. (B) Stabilizing postprandial blood glucose fluctuations and recording levels at different time points using DNJ@MPDA, DNJ@Inulin gel and DNJ@MPDA@Inulin gel and (C) Corresponding AUC calculations. (D) Model of STZ-induced T2DM mice. (E) In 4 weeks, the body weight of mice was recorded. (F) The amount of water-take in 4 weeks. (G) Effect of drugs on Blood Glucose of mice after 4 weeks of treatment. (H) GSP of the five groups mice treated by DNJ, MPDA@Inulin gel or DNJ@MPDA@Inulin gel at same doses for 4 weeks. The data are presented as the mean ± SD (n = 8 biologically independent samples for (B, C, E-H)). Statistical analysis was performed using one-way ANOVA. *p < 0.05, **p < 0.01, and ****p < 0.0001.

2.4. DNJ@MPDA@Inulin gel could effectively modulate glucose metabolism and insulin sensitivity in T2DM mice

Subsequently, the in vivo efficacy of DNJ@MPDA@Inulin gel was assessed in a T2DM mouse model (Fig. 4D). Firstly, T2DM model was established via high fat diet (HFD) for 6 weeks and followed by intraperitoneal injection of streptozotocin (STZ) [35,36]. The successfully induced diabetic mice were randomly divided the into four groups and received the following admistrations: oral administration of PBS (Animal model control group, AC), DNJ (100 mg/kg), MPDA@Inulin gel (MPDA = 2 mg/kg) [37], or DNJ@MPDA@Inulin gel (DNJ = 100 mg/kg, MPDA = 2 mg/kg), respectively. An additional group of healthy mice administered PBS was as a normal control group (NC). The Body weight of the mice was continuously monitored during treatment. According to the Fig. 4E, the body weights of the NC group remained stable, and the other groups exhibited similar trends with no statistically significant differences among them, which potentially attributed to their comparable food intake (Fig. S8). Whereas, the DNJ@MPDA@Inulin gel group exhibited significant reduction in excessive water consumption induced by T2DM (Fig. 4F), demonstrating that DNJ@MPDA@Inulin gel more effectively alleviates classic hyperglycemic symptoms compared to DNJ.

Following the 4-week therapeutic intervention, fasting blood glucose (FBG) levels revealed that mice treated with DNJ@MPDA@Inulin gel displayed significant reductions in both fasting blood glucose levels (Fig. 4G) and glycated serum protein (GSP) levels (Fig. 4H). Moreover, oral glucose tolerance tests (OGTT), which represented overall glycemic control were performed weekly via comparison of glucose levels at all time points with the area under the curve(AUC)calculation. Fig. S9 indicated that glucose tolerance was markedly decreased in T2DM mice, MPDA@Inulin gel exhibited no effect on glucose tolerance. Conversely, in the fourth week, both DNJ and DNJ@MPDA@Inulin gel treatments led to enhanced glucose tolerance, as reflected by reduced AUC values relative to the AC group. Moreover, the therapeutic efficacy increased with administration frequency during the 4-week intervention.

Insulin resistance is the central pathogenic driver of T2DM, serving as a critical biomarker for early risk prediction and targeted intervention. For the insulin tolerance test (ITT) assay, the DNJ and DNJ@MPDA@Inulin gel groups could down regulate the AUC in T2DM mice, DNJ@MPDA@Inulin treated group exhibited the most pronounced reduction (Fig. S10). Fasting serum insulin levels (Fig. S11A) and HOMA-IR analysis (Fig. S11B) also confirmed that DNJ@MPDA@Inulin gel treatment could effectively alleviate insulin resistance. These results indicated that DNJ@MPDA@Inulin gel treatment could effectively ameliorate fasting hyperglycemia, improve glucose tolerance and insulin resistance in T2DM mice.

As a complex disease, persistent hyperglycemia in T2DM can lead to multi-organ impairment. The H&E staining of pancreatic tissues from the NC group exhibited clear boundaries and regular morphology, while the AC group displayed degenerative changes with cytoplasmic vacuoles and nuclear pyknosis. However, DNJ and DNJ@MPDA@Inulin gel groups showed improvement (Fig. 5A). T2DM mice exhibited severe kidney damage, which was partially mitigated by DNJ and DNJ@MPDA@Inulin gel treatment (Fig. 5B). Almost no effect in alleviating kidney damage in MPDA@Inulin treated group was also observed. To further prove the effect of DNJ@MPDA@Inulin in colon tissue, H&E staining, occludin immunohistochemistry and AB-PAS assay were performed. As indicated in Fig. 5C–E, T2DM model demonstrated inflammatory infiltration, epithelial barrier disruption, and goblet cell depletion in colon tissues. DNJ@MPDA@Inulin gel provided superior restoration of intestinal homeostasis compared to DNJ alone, whereas MPDA@Inulin treated group showed negligible effects. Collectively, oral delivery of DNJ@MPDA@Inulin partially reversed glucolipid metabolic dysregulation while suppressing progression of hepatic, renal, pancreatic, and colonic pathologies in T2DM mice. The hydrogel nano-delivery system potentiated the therapeutic impact of DNJ through sustained-release-mediated enhancement of systemic absorption and intestinal retention.

Fig. 5.

The effect of DNJ@MPDA@Inulin on pancreas, kidney and intestine in T2DM mouse. (A) Pancreatic H&E staining (yellow arrows for degeneration of denoted islet cells); (B) Kidney H&E staining (red arrows for tubular vacuolar degeneration; blue arrows for lipid deposition; black arrows for tubular epithelial cell exfoliation; green arrows for tubular brush edge exfoliation). Staining sections of colon (C) H&E staining sections, (D)Immunohistochemical staining sections for Occludin protein and (E) AB-PAS staining of colon tissue. All the samples for section preparation were collected on the 28th day of the gavage experiment. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

2.5. DNJ@MPDA@Inulin gel ameliorates hepatic steatosis

Glycolipid metabolism disturbance mediates insulin resistance and impairs β-cell functionality, driving the progression of T2DM. Inside, increased organ indices of the liver and epididymal fat reflect visceral adiposity and hepatic steatosis, indicating metabolic dysfunction. As shown in Fig. S12, the liver index and epididymal fat index in the AC group was significantly higher than that of the other groups, and DNJ@MPDA@Inulin gel effectively mitigated the increase in both organ indices and inhibited the accumulation of lipids. Furthermore, DNJ@MPDA@Inulin gel mitigated end-organ damage induced by chronic hyperglycemia. Hepatic histology (H&E and Oil Red O staining) revealed structural steatosis, and ballooning degeneration in AC group mice (Fig. 6A and B), pathologies substantially alleviated by DNJ and DNJ@MPDA@Inulin gel treatments though minimally improved by MPDA@Inulin gel alone. Fig. 6C and D revealed that the two-representative liver-function indexes, serum ALT and AST, were reduced following DNJ@MPDA@Inulin gel treatment in T2DM mice. Biochemical analyses further revealed significant reductions in TG, TC, and LDL levels, and the indicator of HDL was improved by DNJ@MPDA@Inulin gel (Fig. 6E–H), corroborating that DNJ@MPDA@Inulin gel possibly affected lipid metabolism. The above results collectively support the conclusion that DNJ@MPDA@Inulin gel has an excellent effect in reducing liver lipid deposition.

Fig. 6.

Amelioration of hepatic steatosis by DNJ@MPDA@Inulin gel in T2DM mice. (A) H&E (Red arrows indicate ballooning degeneration; green arrows mark steatotic lesions) and (B) Oil Red O staining sections of liver. (C-H) Serum biochemical profiles including (C) ALT, (D)AST, (E)TC, (F)TG, (G)LDL, and (H)HDL in normal control (NC), diabetic control (AC), DNJ-treated, and DNJ@MPDA@Inulin gel-treated groups after 4-week intervention. The data are presented as the mean ± SD (n = 6 biologically independent samples for (C–H)). Statistical analysis was performed using one-way ANOVA. p > 0.05 (n.s.), *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

2.6. DNJ@MPDA@Inulin gel reprogrammed the gut microbiota structure

Gut microbiota alterations contributed to symptom amelioration in T2DM mice. Parameters including the Chao and observed species suggested that DNJ@MPDA@Inulin gel administration moderately enhanced the microbial abundance and diversity in the gut of T2DM mice (Fig. 7A and B). Principal Coordinates Analysis (PCoA) based on Bray-Curtis dissimilarity metrics revealed that there were relative differences in bacterial communities between the tested four groups (Fig. 7C). Furthermore, Non-metric Multidimensional Scaling (NMDS) analysis revealed that the gut microbiota composition of DNJ@MPDA@Inulin gel group was the closest to that of the NC group, indicating the highest degree of similarity between these two groups (Fig. S13). Stacked bar charts of microbial abundance at the phylum (Fig. S14) and genus (Fig. 7D) levels revealed pronounced compositional shifts between NC and AC groups, confirming that T2DM markedly perturbed the gut microbiota. Both DNJ and DNJ@MPDA@Inulin gel treatments further reshaped the microbiota of T2DM mice. At the phylum level, T2DM elevated the Firmicutes/Bacteroidetes ratio (Fig. 7E), and although this ratio in the DNJ@MPDA@Inulin gel group did not fully revert to the NC group, it displayed a clear trend toward restoration. Notably, both DNJ and DNJ@MPDA@Inulin gel significantly enriched beneficial genera including Akkermansia, Alloprevotella, and Alistipes (Fig. 7D). These genera are known to be associated with short-chain fatty acid (SCFA) production, and contribute to improved intestinal barrier function, bile acid metabolism, and anti-inflammatory effects.

Fig. 7.

Gut microbiota remodeling by DNJ@MPDA@Inulin gel in T2DM mice. Alpha diversity violin plots of the gut microbiota of the mice of four groups. (A) Chao index; (B) Observed_features. (C) Principal Coordinates Analysis (PCoA) derived from Bray−Curtis distance. (D) The profile of gut microbial community at the genus level (Top 20 most abundant genera). (E) Violin plot of the Firmicutes-to-Bacteroidetes ratio. (F) Venn diagram of gut microbiota between NC-AC, AC-DNJ, and AC-DNJ@MPDA@Inulin gel groups, with values between different circles representing the number of the same species. (G) Heatmap depicting the correlation between the specific gut microbiota (filtered by Venn analysis) and serum biochemical indices. Relative abundance of (H) Coriobacteriaceae_UCG-002, (I) Akkermansia, (J) Ruminococus, and (K) Prevotellaceae_UCG-001 collected from (G). The data are presented as the mean ± SD (n = 5); Statistical significance was determined using two-tailed Student's t-tests.

Meta-analysis of differentially abundant genera among NC VS AC, AC VS DNJ, and AC VS DNJ@MPDA@Inulin gel groups revealed treatment-specific microbial shifts (Venn diagram, Fig. 7F). Correlation analyses were conducted to quantify these relationships, with the results visualized in Fig. 7G. Notably, the abundance of the harmful genus Coriobacteriaceae_UCG-002 (Fig. 7H) [38] and Adlercreutzia (Fig. S15A) [39] decreased. In contrast, beneficial genera such as Akkermansia (Fig. 7I) [40] and Allobaculum (Fig. S15B) [41] showed increases, both of which are known to enhance intestinal barrier integrity, exert anti-inflammatory effects, and contribute to SCFA production. These findings are consistent with previous studies on the preventive and therapeutic effects of DNJ in type 2 diabetes mellitus (T2DM) [42]. Moreover, the DNJ@MPDA@Inulin gel group exhibited a marked increase in Ruminococcus (Fig. 7G) [43], a genus involved in complex carbohydrate degradation. Additionally, inulin-utilizing taxa, including Prevotellaceae_UCG-001 (Fig. 7K) [44] and Bifidobacterium (Fig. S15C) [45], were notably enriched in this group. Collectively, these findings suggested that inulin degradation facilitated the proliferation of multiple fiber-degrading probiotics, enabling the DNJ@MPDA@Inulin gel treatment to restore gut microbial balance more effectively than free DNJ administration.

3. Conclusions

In summary, we successfully developed a dual-component strategy where MPDA immobilizes DNJ through hydrogen bonding and pore adsorption, and the inulin gel encapsulates the DNJ@MPDA complex. This inulin hydrogel remains structurally stable in the gastric acid environment, protecting the DNJ@MPDA complex during gastric transit, thereby minimizing the burst release of DNJ. Oral delivery of DNJ via MPDA@Inulin gel significantly prolonged intestinal retention and maintained therapeutic concentrations through sustained release, demonstrating enhanced bioavailability. Long-term oral administration of DNJ@MPDA@Inulin gel could effectively maintain blood glucose levels and improve insulin sensitivity. Notably, it was found that this hydrogel nano-platform based DNJ delivery system also regulates hepatic glucose and fatty acid metabolism, repair intestinal barrier function and modulate gut microbiota. Despite these promising results, the MPDA@inulin gel system, while demonstrating excellent sustained release of DNJ, its clinical translation is hindered by two critical challenges. First, industrialization barriers—including low MPDA yield, inconsistent batch quality upon scale-up, and undefined storage stability—compromise its economic feasibility. Second, interspecies differences in gut microbiota may alter carrier degradation and DNJ release kinetics, potentially affecting therapeutic outcomes. Therefore, future efforts will focus on optimizing scale-up production and validating the system in preclinical models with greater translational relevance to address these limitations. Nevertheless, owing to the sustained-release and microbiota modulation capacity, we believe that inulin-MPDA based biomimetic system holds promise as an ideal candidate for oral bioactive substance delivery and broad biomedical applications.

3.1. Materials and methods

Materials. DNJ(C₆H₁₃NO₄) was purified from mulberry leaves by a liquid chromatography-mass spectrometry (LC-MS) system (Waters, USA) as previously reported¹¹, its purity was 98% by high-performance liquid chromatography (HPLC) analysis. Dopamine hydrochloride (C₈H₁₁NO₂·HCl, 98%) was obtained from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Pluronic® F127 (Poloxamer 407) was purchased from Beyotime Biotechnology Co., Ltd. (Shanghai, China). Streptozotocin (STZ, C₈H₁₅N₃O₇) was supplied by Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China). Ammonium hydroxide (NH₄OH), 1,3,5-trimethylbenzene (TMB, C₉H₁₂), and absolute ethanol (C₂H₅OH) were acquired from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All chemicals were of analytical grade, and ultrapure water was used for all experimental procedures.

3.2. Preparation of MPDA and DNJ@MPDA

MPDA was prepared using the conventional soft-template method [46], 0.05 g of F127 and 0.025 g of DA were dissolved in 5 mL of water/ethanol (1:1, v/v), stirred at room temperature until clear. Then, 0.075 mL of TMB was added with stirring at 500 rpm to form a nanoemulsion. After 30 min, 0.25 mL of concentrated NH₄OH was dropwise added to induce dopamine oligomer self-polymerization. After another 30 min of reaction, the polymer nanospheres were centrifuged and washed sequentially with water and acetone/ethanol (1:3, v/v) three times. DNJ was then added into the MPDA solution and stirred for 24 h [47].

3.3. Characterization of MPDA and DNJ@MPDA

TEM (Tecnai G2F30 S-Twin, TheNetherlands) and SEM (Nova NanoSEM 450, FEI, America) were used to analyze their morphologies. The Malvern laser particle size analyzer (Zetasizer 3000, Britain) determined the size and ζ potential of MPDA and DNJ@MPDA nanoparticles [48]. UV-vis spectra were obtained with a UV-1900i spectrophotometer (Shimadzu, Japan).

3.4. α-Glucosidase inhibition assay measurements

First, 35 μL of 100 mM Tris-maleate buffer was added to the test (A1) and blank control (K1) groups, while 40 μL was added to the test background (A) and blank background (K2) groups. Next, 5 μL of DNJ solution (diluted to approximately 80 μM) was dispensed into the A1 and A groups, whereas 5 μL of ultrapure water was added to the K1 and K2 groups as replacement. Then, 5 μL of α-glucosidase solution (4 U/mL) was introduced to the A1 and K1 groups, while the A and K2 groups received no enzyme. The reaction mixtures were incubated at 37 °C for 15 min before adding 5 μL of maltose substrate solution (0.1 M) to all groups. After another 1-h incubation at 37 °C, the reactions were terminated by adding 50 μL of Tris-maleate-NaOH buffer (pH 8.0). Glucose production was quantified using a glucose assay kit (Nanjing Jiancheng Bioengineering Research Institute, China) by measuring absorbance. The absorbance detected by the corresponding groups was calculated using the formula to obtain the inhibition rate:

$$\text{Inhibition}(\%)=\left(1-\frac{\left(A1-1\right)}{\left(K1-K2\right)}\right)\times 100\%$$

3.5. Encapsulation efficiency (EE) measurements

The DNJ@MPDA mixture underwent centrifugation at 12000 rpm for 10 min at 4 °C for supernatant collection. The collected supernatant was diluted 200 times for unencapsulated DNJ quantification via HPLC. Each experiment was performed three times. Here, W0 denotes the initial DNJ amount in the formulation, while W represents the unencapsulated DNJ in the supernatant. DNJ encapsulation efficiency (EE%) was calculated using the equation:

$$EE(\%)=\left(1-\frac{W}{W0}\right)\times 100\%$$

3.6. Synthesis of DNJ@MPDA@Inulin gel

3 mL water and 2.16 g of inulin were mixed and stirred to achieve uniformity. Subsequently, heated the mixture at 70 °C with stirring at 600 rpm for 6 min. Then, 2.5 mL of heated inulin solution and 0.5 mL of pre - prepared DNJ@MPDA solution were quickly mixed thoroughly. Finally, the mixture was left to stand at 4 °C for 12 h to form the DNJ@MPDA@Inulin gel. MPDA@inulin gel was also prepared using the same method. The formulation consists of 37.5 wt% and 27.5 wt% of base inulin gel was prepared by mixing 0.6 g and 0.3 g of inulin with 1 ml of water and then following the heating method mentioned above.

3.7. Characterization of DNJ@MPDA@Inulin gel

The structural and rheological properties of DNJ@MPDA@Inulin gel and inulin-based control gels were systematically characterized through SEM analysis of freeze-dried samples and comprehensive rheological evaluation using a rheometer (DHR-2, TA, USA) with 25-mm parallel plates. Viscoelastic behavior was determined via frequency sweep tests at 0.1% constant strain over an angular frequency range of 0.1-100 rad/s, while flow sweep measurements at shear rates of 0.1-100 s⁻¹ characterized the shear-thinning properties of the hydrogel systems.

3.8. In vitro degradation analysis of DNJ@MPDA@Inulin gel

Prior to gelation, 1 mL aliquots of DNJ@MPDA@Inulin gel solution were transferred to centrifuge tubes. Initial measurements recorded empty tube mass (m₀) and gel-containing tube mass (m₁). Subsequent incubation proceeded at 37 °C in SGF, SIF, or PBS buffers. At designated intervals, removal of supernatant followed by tube reweighing (m) enabled percentage mass loss calculation, with triplicate measurements conducted for each time point.

$$W\%=\frac{m-m0}{m1-m0}\times 100\%$$

3.9. Stability of DNJ@MPDA@Inulin gel at different physiological simulated fluids

The DNJ@MPDA@Inulin gel and other Inulin base gels were immersed in SGF, SIF, and PBS for 6 h, and their structural integrity was determined using SEM. Simulated gastric fluid (SGF, pH 1.5) was prepared with 0.2% NaCl and 0.16% pepsin in deionized water, then adjusted with HCl. Simulated intestinal fluid (SIF, pH 6.8) contained 0.68% K₂HPO₄ and 1% pancreatin in aqueous solution, with pH adjusted using NaOH/HCl.

3.10. In vitro release of DNJ from DNJ@MPDA@Inulin gel

Hydrochloric acid (HCl) and sodium hydroxide (NaOH) were used to regulate the acidity and basicity of the PBS. In this experiment, PBS with varying pH values (pH = 1.5, 6.0, 7.4, and 8.0) was used. DNJ solution and DNJ@MPDA@Inulin gel were loaded into dialysis bags (molecular weight cutoff: 3500 Da) and incubated in an orbital shaker at 37 °C and 100 rpm. Over time, samples were taken after adding corresponding volumes of PBS buffer. The concentration of DNJ was determined by HPLC.

3.11. Retention of MPDA@Inulin gel in the gastrointestinal system

For each milliliter of MPDA nanoparticles, 0.75 mg of Cy7 was added, and the mixture was incubated in an orbital shaker for 24 h. The resulting Cy7-MPDA conjugate was then used to prepare Cy7-MPDA@Inulin gel following the aforementioned procedure. Mice received Cy7-MPDA nanoparticles or Cy7-MPDA@Inulin gel via oral gavage. They were dissected at 2, 4, 6 h post-administration, and the stomach - to - colon tissues were imaged with the IVIS system (Ex = 649 nm, Em = 670 nm).

3.12. Animals

SPF-grade male C57BL/6 mice (4-6 weeks old) were acquired from the China National Laboratory Animal Center (Shanghai, China) and maintained under controlled conditions with free access to feed and water during a 7-day acclimation period. All experimental protocols received approval from the Zhejiang University of Technology's Institutional Animal Care and Use Committee.

3.13. The immediate hyperglycemia model induction and blood glucose monitoring

Previous to the experiment, mice were fasted for 12 h, and fasting blood glucose (FBG, 0 h) was measured using a glucometer (Sinocare, China). The mice were then randomly divided into three groups (6 per group) based on FBG levels. Initially, all mice received an oral gavage of sucrose solution (6 g/kg, 200 μL). Immediately after sucrose administration, treatments were given via oral gavage: normal control (NC, equal volume of PBS), DNJ (100 mg/kg), and DNJ@MPDA@Inulin (100 mg/kg) groups. To sustain elevated blood glucose, a second equal dose of sucrose was administered 3 h post initial sucrose load. Blood glucose levels were monitored at 0.25, 0.5, 1, 1.5, 2, 3, 3.25, 3.5, 4, 4.5, 5, and 6 h after the first sucrose dose using tail vein blood sampling. The area under the glucose tolerance curve (AUC) was calculated utilizing GraphPad Prism 9.10 software for quantitative analysis.

3.14. T2DM model induction and remission

The T2DM model was established via HFD feeding combined with low-dose STZ injection, a well-characterized protocol that mimics clinical pathogenesis by sequentially inducing insulin resistance and partial β-cell dysfunction. This approach preserves residual β-cell function with high stability and low mortality, allowing for comprehensive evaluation of drug efficacy [35]. 48 mice were randomly divided into a control group (n = 8, normal control, NC) and a T2DM model group (n = 40). Control mice received standard chow, while model mice were fed a high-fat diet (HFD; D12492, 60% kcal from fat) for 4 weeks. After a 16-h fast, model mice received daily intraperitoneal injections of streptozotocin (STZ, 50 mg/kg) for five consecutive days, whereas control mice received citrate buffer (0.1 mol/L, pH 4.2). 7, 14, 21 days post-STZ administration, FBG was measured via tail vein sampling. T2DM was confirmed if FBG levels exceeded 11.0 mmol/L in three consecutive measurements. Successfully induced diabetic mice were further randomized into four groups (n = 8/group): diabetic model (AC), DNJ (100 mg/kg/d), MPDA@Inulin (1.25 mg/kg/d), and DNJ@MPDA@Inulin (100 mg/kg/d) groups. Throughout the study, control mice remained on standard chow while other groups continued HFD feeding.

3.15. Oral glucose tolerance test (OGTT)

Before commencing the experimental procedures, all mice underwent a 16-h fasting period followed by oral administration of glucose (2 g/kg). Venous blood samples were obtained via tail puncture at specified time points (0,15,30,60 and 120 min post-administration), with blood glucose concentrations determined using a commercial glucometer. The AUC was calculated utilizing GraphPad Prism 9.1.0 software for quantitative analysis of OGTT results.

3.16. Insulin tolerance test (ITT)

All mice were subjected to a 4-h fast before an intraperitoneal insulin injection (0.5 U/kg). Blood samples were collected and analyzed as in the OGTT.

3.17. Serum biochemical index test

AST, ALT, TG, TC, HDL, LDL, GST and glucose levels were detected using the corresponding test kits (Nanjing Jiancheng Bioengineering Research Institute, China). Insulin levels were analyzed using a mice insulin ELISA kit (Jingmei Biotechnology Co., Ltd., Jiangsu, China) according to the instructions provided in the kit manual.

3.18. Histopathology studies

At the experimental endpoint, all tissue specimens were harvested for comprehensive histological assessment. Hepatic, colonic, renal, and pancreatic morphology was examined through hematoxylin and eosin (H&E) staining [49], while hepatic steatosis was quantified via Oil Red O staining. Intestinal barrier integrity was assessed by immunohistochemical detection of occludin expression in colonic tissues. Goblet cell distribution and mucin production were evaluated using Alcian Blue/Periodic Acid-Schiff (AB/PAS) staining. All histopathological analyses were performed using standard light microscopy techniques.

3.19. Statistical analyses

Data from at least three independent experiments (n ≥ 3) were presented as mean ± SD. Two-sample t-tests or one-way ANOVA were used for statistical comparisons between pairs or multiple groups, respectively. Statistical significance was defined as p > 0.05 (n.s.), *p < 0.01, **p < 0.001, and ***p < 0.0001.

4. Associated content

The MPDA concentration and absorbance Linear Fit; Inhibition curve of α -glucosidase by DNJ; FTIR spectrum and Zeta potential before and after DNJ@MPDA@Inulin gel; Synthesis and characterization of 27.5 wt% Inulin gel; Food intake and GSP of the five groups mice after 4 weeks; The OGTT and AUC of OGTT during the first three weeks of DNJ@MPDA@Inulin gel treatment; Fasting insulin and HOMA-IR histogram; The liver and the epididymal organ index; Beta diversity plot of the gut microbiota of NMDS and PCoA plot; The profile of gut microbial community at the phylum level.

CRediT authorship contribution statement

Shi Zhong: Investigation, Writing – original draft. Lulu Qi: Investigation, Writing – original draft. Jinxi Huo: Investigation. Yunqing Sun: Investigation. Jiayi Feng: Investigation. Yuanxiang Jin: Formal analysis, Investigation. Hangjun Chen: Formal analysis, Investigation. Qinglian Hu: Funding acquisition, Project administration, Writing – review & editing. Yougui Li: Formal analysis, Funding acquisition, Project administration, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.