Mannose coated selenium nanoparticles normalize intestinal homeostasis in mice and mitigate colitis by inhibiting NF-κB activation and enhancing glutathione peroxidase expression

Abstract

Impaired intestinal homeostasis is a major pathological feature of inflammatory bowel diseases (IBD). Mannose and selenium (Se) both demonstrate potential anti-inflammatory and anti-oxidative properties. However, most lectin receptors bind free monosaccharide ligands with relatively low affinity and most Se species induce side effects beyond a very narrow range of dosage. This has contributed to a poorly explored therapies for IBD that combine mannose and Se to target intestinal epithelial cells (IECs) for normalization gut homeostasis. Herein, a facile and safe strategy for ulcerative colitis (UC) treatment was developed using optimized, mannose-functionalized Se nanoparticles (M-SeNPs) encapsulated within a colon-targeted hydrogel delivery system containing alginate (SA) and chitosan (CS). This biocompatible nanosystem was efficiently taken up by IECs and led to increased expression of Se-dependent glutathione peroxidases (GPXs), thereby modulating IECs’ immune response. Using a mouse model of DSS-induced colitis, (CS/SA)-embedding M-SeNPs (C/S-MSe) were found to mitigate oxidative stress and inflammation through the inhibition of the NF-kB pathway in the colon. This stabilized mucosal homeostasis of IECs and ameliorated colitis-related symptoms, thereby providing a potential new approach for treatment of IBD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-024-02861-2.

Introduction

Ulcerative colitis (UC) is a type of inflammatory bowel disease (IBD) that presents as a chronic, relapsing-remitting inflammatory condition initiated proximal to the colon with a high incidence and prevalence globally [1, 2]. Emerging evidence has suggested a crucial role of the intestinal epithelium cells (IECs) in the development and persistence of UC [3–6]. In addition to their barrier role, IECs exhibit inherent immunoregulatory capabilities that significantly contribute to actively combating pathogens [6–8]. During pathogenic challenge that triggers proinflammatory cytokine secretion by immune and stromal cells, IECs themselves secrete TNF-α, interleukin (IL)-1β, IL-6 and MCP-1. This response is primarily regulated by the nuclear factor (NF)-κB signaling pathway as a key modulator of intestinal immune homeostasis [6]. In addition, innate immunity driven by IECs also involves the release of reactive metabolites of oxygen in the luminal direction to increase oxidative stress that sustains chronic inflammation. Current investigations into the pathophysiology of IBD have predominantly concentrated on mechanisms mediated by infiltrating immune cells [9, 10], while limited research has been dedicated to therapeutic interventions targeting IECs. Preferred treatments for UC include anti-inflammatory drugs such as salazosulfa pyridine and 5-aminosalicylic acid compounds with inherent issues including inconsistent efficacy and undesirable side effects [11, 12]. This emphasizes the urgent need to find new strategies for UC treatment.

Specific binding of lectin to ligands on IECs plays a pivotal role in regulating intestinal homeostasis and inflammatory processes driving IBD, which involve interactions between mannose and the mannose receptor - CD206 (M-MRs, CD206 is a C-type lectin receptor) [13, 14]. Mannose is a simple hexose sugar that has been shown to possess anti-inflammatory effects [15–20]. Data supports the beneficial effects of mannose in IBD through the modulation of the inflammatory state of IECs [21, 22]. For example, Zuo et al. showed that high concentrations of mannose supplementation could reduce UC symptoms by attenuating intestinal barrier damage [23, 24]. A separate study proposed that mannose decreased IL-1β production and thereby suppresses lipopolysaccharide (LPS)-induced inflammation in the gut [19]. The majority of lectin receptors exhibit a low affinity for monosaccharide binding, however, which limits the anti-inflammatory efficacy of free mannose [25]. To overcome the limitations of low receptor affinity and expression levels, mannose may be coated onto nano materials [26–28]. Our previous study supported a reduction of drug dosage by using a nanoparticle (NP) approach, and a particularly effective combination of mannose-rich oligosaccharide (MRO) coated selenium NPs (SeNPs) leading to macrophage reprogramming and inflammatory resolution that reduced the severity of UC [29, 30]. These effects can be attributed to the larger specific surface area of NPs and their capability to induce clustering of cell surface receptors [27, 28].

Se is an essential nutritional trace element incorporated into selenoproteins that play a wide range of biological roles including the regulation of oxidative stress and cellular redox tone [31, 32]. Se deficiency has been considered a risk factor for exacerbating intestinal injury and promoting inflammatory responses in UC [33–35]. This is mainly attributed to decreased bioactivity of antioxidant selenoproteins including glutathione peroxidases (GPXs), which are critical for reducing peroxides that can exacerbate disease [36]. In cases of nutritional deficiency, Se supplementation can be beneficial for individuals suffering from UC. In addition, the substantial reduction in the Se uptake by damaged IECs during IBD necessitates a higher level of Se nutritional supplementation. Compared with other forms of Se such as selenate (SeO₄ 2-, + 6), selenite (SeO₃ 2-, + 4) and selenide (H₂Se, − 2), elemental Se (0) present in SeNPs are less toxic and more biocompatible [37–40]. SeNPs offer versatility with optimizable size, surface charge, and drug loading capacity with controlled release [41, 42]. For these reasons, SeNPs have emerged as a potential therapeutic modality for UC treatment [43]. Combining the anti-inflammatory effects of mannose with this modality in the form of a mannose functionalized SeNPs (M-SeNPs) may enhance the modulation of IECs in UC and provide an effective therapeutic intervention strategy.

For the current study, we report of a novel nanosystem based on M-SeNPs modified by ascorbic acid reduction of Na₂SeO₃ with mannose supplemention. Optimized M-SeNPs were physically characterized for size, stability, morphology, loading capacity and chemical structure. Using an established mouse model of DSS-induced colitis, we embedded M-SeNPs in a colon-targeted hydrogel CS/SA to form C/S-MSe microbeads and assessed the capacity of M-SeNPs to target and interact with IECs, thereby increasing the expression of GPXs in the colon. Our results demonstrate that these modified M-SeNPs counteracted the oxidative stress and inflammation that is found in colitis lesions and alleviated the pathology driving UC.

Materials and methods

Preparation and characterization of M-SeNPs

M-SeNPs were prepared by supplementing ascorbic acid into a mixture of mannose and Na₂SeO₃ for reduction as previously described [30], with slight modifications. Briefly, 100 mg mannose was dissolved in 10 mL deionized water (10 mg/mL), and 500 µL of Na₂SeO₃ (10 mM) was mixed with different volumes of mannose solution to reach final mannose concentrations of 0.125, 0.25, 0.5, 0.75 and 1 mg/mL. These mixtures were vigorously stirred for 0.5 h at room temperature until completely dissolved. Stirring speed was slowed down to approximately 350 rpm, then 5 mL of ascorbic acid solution (4 mM) was added drop by drop into the mixture at 4℃ for 12 h in dark conditions. After dialysis for another 24 h against Milli-Q water to remove excess ascorbic acid and Na₂SeO₃, a solution of modified M-SeNPs were obtained. Mannose-negative control SeNPs were generated by replacing the mannose solution with the same volume of deionized water during the preparation of M-SeNPs. M-SeNPs&C6 were prepared by an added step of 20 µL C6 fluorescence (1 mg/mL resolved in ethanol) into the scheme before reduction. Note that the concentration of mannose was the same as the mannose concentration loaded on coherent M-SeNPs calculated by the loading rate prepared as described above. The loading rate of mannose onto M-SeNPs was detected using a phenol sulfuric acid method before/after dialysis.

The size distribution and zeta potential of SeNPs and M-SeNPs were measured by dynamic light scattering (DLS) using a Zetasizer Nano ZS instrument (Malvern Instruments, Malvern, UK). Morphological characterization of SeNPs and M-SeNPs was examined by transmission electron microscopy (TEM, Hitachi H-7650, 100 kV), with energy dispersive X-ray (EDS) coupled with TEM for element quantification & mapping of M-SeNPs. Fluorescence spectrum of C6 labeled SeNPs (SeNPs&C6) and M-SeNPs (M-SeNPs&C6) was analyzed by Lumina Fluorescence (Thermo Fisher Scientific, USA). The UV spectrum of the SeNPs and M-SeNPs were monitored in the range of 200–900 nm by a UV-Vis spectrophotometer (UV-2550, SHIMADZU, CHN), and FT-IR (Equinox 55 IR spectrometer, Thermo Fisher Scientific, USA), with these analyses used to determine the infrared absorption spectrum of M-SeNPs.

Microfluidic fabrication of C/S-MSe

The colon-targeted hydrogel embedded M-SeNPs (C/S-MSe) were synthesized by a microfluidics approach optimized based on a previously described method [44]. Briefly, disperse M-SeNPs in the 1% sodium alginate (SA) to form the dispersed phase, then place it in a microfluidic device. 2% droplet generation oil was used as the continuous phase to collect the resulting microdroplets. Then remove the continuous phase and wash multiple times with pure water, immerse them in a CS solution (1%, w/v), with continuous shaking for 30 min (100 rpm, 37 °C), forming hydrogel microbeads C/S-MSe. The morphology and size distribution of microbeads were obtained from the imageJ statistical microscope microbeads photographs. The microstructure of C/S-MSe was observed using a field emission scanning electron microscope (SEM, Hitachi, Japan, S4800), and the trace element mapping were obtained by an EDS (Thermo Science, ESCALAB 250Xi). The cumulative rate of M-SeNPs in pH 1.2 and pH 6.8 buffers that simulate the pH of gastric and intestinal fluid was tested by a UV-Vis spectrophotometer.

Cells and stimulation conditions

Human colonic epithelial NCM460 cells and human intestinal epithelial HIEC cells were purchased from the Cell Bank of Chinese Academy of Sciences. NCM460 cells and HIECs were cultured in RPMI 1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 100 U/mL penicillin, 100 µg/mL streptomycin, and 10% fetal bovine serum (FBS). To induce inflammation in IECs, NCM460 cells were seeded at a density of 2 × 10⁵ cells/well in a 6-well plate and incubated overnight at 37℃ with 5% CO₂ followed by LPS (100 ng/mL) for 6 or12 h for mRNA and protein isolation, respectively. M-SeNPs (1 µM) were pre-treated before LPS for 18 h. To examine the effect of M-SeNPs on DSS-induced injury in IECs, NCM460 cells were seeded at a density of 1 × 10⁴ cells/well in a 96-well plate. NCM460 cells were then stimulated with 30 mg/mL of DSS for 12 h and 1 µM M-SeNPs for another 12 h.

Cellular uptake of M-SeNPs in NCM460

NCM460 cells were seeded overnight at a density of 5 × 10⁴ cells/well in a 24-well plate and 1 µM of coumarin-6 (C6) labeled M-SeNPs (M-SeNPs&C6) added to trace the intracellular location for different times (0, 0.5, 3, 6, 10 h). Subsequently, cells were washed 3 times with PBS to remove the residual fluorescence. NCM460 cells were then incubated with 37℃ pre-warmed complete medium containing 50 nM of Lysotracker (Invitrogen, L7528) for 30 min, following 5 min complete medium incubating for 3 times at 37℃ in 5% CO₂-containing incubator. Cells were then counterstained with 4′, 6-diamidino-2-phenylindole (DAPI) for 10 min. Confocal images of the co-localization of SeNPs&C6 and M-SeNPs&C6 with lysosome was observed by using the Olympus (FV10).

Colitis model induction and animal treatment: Male C57BL/6 mice (6-wk-old) used in this study were supplied by the Sun Yet-Sen University Animal Breeding Unit (Guangzhou, China). All animal care and experimental protocols were conducted under the approval of the Animal Care and Ethics Committee of Jinan University (20210725-04). Animals were housed in mouse cages (M4 in size of 370 mm × 260 mm × 170 mm), under standard conditions of temperature (21 ± 3℃), light (12 h light/dark cycle), and humidity (55 ± 15%), with ad libitum access to water and food. The mice were adapted for 1 wk before conducting experiments, randomly divided into 5 groups: control, DSS, mannose (DSS + 0.5 ppm mannose that calculated by the loading rate of mannose on SeNPs), C/S-Se (DSS + 1 ppm SeNPs embedded in CS/SA hydrogel) and C/S-MSe group (DSS + 1 ppm M-SeNPs embedded in CS/SA hydrogel). For the colitis model, mice were given ad libitum of 3.0% (w/v) DSS-containing water for 7 d, with different formulations administered during the induction period via gavaging. Body weights were recorded daily with weight changes recorded relative to the initial weight. All mice were euthanized at day 14 for tissues and blood collection. Plasma samples were collected by blood centrifuging at 3000 rpm for 10 min and stored at -80 ℃. Fresh colons were excised with the scalpel, including the residual tube starting from the caecum collected as proximal colon, followed by the distal colon and middle colon. Segments were further cut into 3–4 mm pieces and placed in 4% PFA for histological assessment or freshly stored at -80℃.

Disease activity index (DAI)

The DAI was evaluated by scoring the clinical manifestations daily, including body weight, stool consistency and rectal bleeding [45] as described previously by Zhang et al. [46], and the scoring system is shown in Table 1. Weight loss represents the weight change relative to the previous day, and diarrhea was characterized by the absence of fecal pellet formation and the presence of continuous fluid fecal material in the colon. Occult blood was assessed in fecal using a kit, and the values of DAI were calculated by summing the score of weight loss, diarrhea and rectal bleeding.

Table 1.

Scoring criteria for DAI of experimental colitis in mice

Weight loss (%)	Stool consistency	Occult blood	Score
<0	Normal	Negative	0
1–5		+	1
5–10	Loose	++	2
10–20		+++	3
>20	Diarrhea	Gross bleeding	4

Histopathology analysis

Mice were sacrificed at the end of the experiment with entire colons carefully dissected and separated into 3 segments including the proximal colon, middle colon and distal colon. These were washed with cold PBS, then fixed in 4% (v/v) buffered PFA and embedded in paraffin to provide sections for histological evaluation. The histological colon sections were cut into 5 μm thick slices and immobilized on slides for hematoxylin and eosin (H&E) staining, and blindly scored as previously described [47].

Determination of total Se

All tissues including colon (proximal, middle and distal), liver, kidney, spleen, and small intestine were extracted and carefully weighed. Tissues were then mineralized in ICP-MS grade HNO₃ (65%), and diluted 100-fold in water. The concentration of total Se was determined by inductively coupled plasma mass spectrometry (ICP-MS) as described previously [48].

Statistics analysis

Data were all analyzed using Prism software (version 9.01; GraphPad Software) and presented as mean ± standard deviation (SD). Comparisons of groups were made using one-way ANOVA. A P-value < 0.05 (*) was considered statistically significant.

Results

Preparation and characterization of M-SeNPs

Although mannose supplementation has been shown to exert positive effects to colitis by maintaining tight junctions in IECs, mannose in circulation may exert other effects [23]. Our previous work indicated that SeNPs could reduce the effective dosage of loading ligand [30]. For this study, M-SeNPs in the context of DSS induced colitis mice was analyzed for effects on UC symptoms and potential mechanisms involving IECs. M-SeNPs were prepared with SeNPs used as a control, which both were characterized and M-SeNPs found to exhibit a transparent appearance with the smallest size of diameter = 106.4 ± 2.1 nm (Figure S1A-C). As shown in Fig. 1A-B, the control SeNPs were aggregated to a greater diameter of 300 nm, while the M-SeNPs were monodispersed, homogeneous and spherical, with a relatively narrow diameter distribution. Zeta potentials of SeNPs and M-SeNPs were measured at -10.9 ± 0.6 and − 12.4 ± 0.5, indicating the electrostatic interaction and a higher stability of M-SeNPs (Figure S1D). The elemental composition of M-SeNPs was determined by energy dispersive X-ray (EDX). Figure S1E & Fig. 1C shows the atomic percentage of Se (41.62%), C (51,75%) and O (6.63%) in M-SeNPs, indicating an effective loading of mannose onto M-SeNPs. For a higher resolution characterization, M-SeNPs were encapsulated with C6 [49–51], and the loading rate of mannose was detected by phenol sulfuric acid method. As shown in Figure S1F, 26.0% and 41.7% of mannose were loaded on M-SeNPs&C6 and M-SeNPs respectively. Accordingly, the molar ratio of mannose to Se in the M-SeNPs was 1:25. Results of UV-vis spectrum analysis for mannose, SeNPs, and M-SeNPs showed an absorbance peak at 266 nm for SeNPs, and the addition of mannose did not change the peak position (Figure S1G). Moreover, the fluorescence spectra of SeNPs, SeNPs&C6, M-SeNPs and M-SeNPs&C6 were scanned as shown in Figure S1H, demonstrating successful addition of C6 with a peak at 517 nm. The results of Fourier transform infrared (FT-IR) spectroscopy showed a characteristic peak of mannose at ~ 810/870 cm^− 1 that was also present in M-SeNPs. A significant peak at 1143.44 cm^− 1 observed in M-SeNPs was attributed to Se = O asymmetric stretching, which further confirms the successful incorporation of mannose into M-SeNPs (Fig. 1D). By using microfluidic technology, M-SeNPs were embedded into CS/SA hydrogel microbeads to create C/S-MSe hydrogel microbeads. As illustrated in Fig. 1E, the cumulative release of M-SeNPs from C/S-MSe at 24 h in a pH 6.8 buffer (46.1%), which mimics intestinal fluid, is significantly higher compared to the release in a pH 1.2 buffer (19.8%) that simulates gastric fluid. This results indicates that C/S-MSe is pH-sensitive, thereby aiding in its colon-targeting capability. Moreover, Fig. 1F-G reveal that the microbeads produced via microfluidic control exhibit a uniform size distribution, with the M-SeNPs loadded microbeads displaying a light orange hue and an average particle size of approximately 169 μm. Overall, these data demonstrate the successful synthesis of M-SeNPs and C/S-MSe with appropriate physiochemical properties.

Fig. 1.

Preparation and characterization of the M-SeNPs. (A) The size distribution profiles and appearance of SeNPs (top) and M-SeNPs (bottom). (B) TEM images of SeNPs (top row) and M-SeNPs (bottom row). (C) STEM-EDS elemental analysis of the M-SeNPs. (D) FT-IR spectra of Mannose, SeNPs and M-SeNPs. (E) Cumulative release of M-SeNPs under pH 1.2/6.8. (F) Micrograph and (G) size distribution of microdroplets produced by chitosan/sodium alginate (CS/SA) microbeads in a microfluidic generator with (bottom) or without (top) M-SeNPs embedded. (H) Element mapping of C/S-MSe microbeads

Biocompatible M-SeNPs protected NCM460 cells against DSS-Induced Injury

Efficient cellular uptake and biocompatibility are considered a prerequisite for the therapeutic efficacy of M-SeNPs. To study the uptake of M-SeNPs in IECs, C6 fluorescence labeled M-SeNPs (M-SeNPs&C6) were constructed and aggregation within cells was assessed. Colocalization experiments were performed by confocal microscopy with the colocalization of M-SeNPs with the lysosomal marker indicating that M-SeNPs were localized to the lysosome. Additionally, the uptake of M-SeNPs was higher compared to bare SeNPs, indicating improved biocompatibility of M-SeNPs, which may be attributed to the mannose modification. (Fig. 2A). Furthermore, a DSS-induced injury model was established in NCM460 to determine the effects of M-SeNPs on IECs. This model involved NCM460 cells exposed to 20 µM DSS and treated with different concentrations (0.25, 0.5, 1, 2, 4 µM) of mannose, Na₂SeO₃, SeNPs and M-SeNPs for 12 h. Results showed a significant decrease of cell viability below 68% after of DSS exposure, with only M-SeNPs exerting protective effects in a dose-dependent manner (Fig. 2B & Figure S2A). These data indicate that the M-SeNPs could be taken up into lysosome of IECs through endocytosis with high compatibility that led to a protective effect in IECs from DSS-induced injury (Fig. 2C). Biocompatibility of M-SeNPs was investigated by using CCK-8 assay in colonic relevant cell lines such as NCM460, peritoneal macrophages (PMs) and HIEC. The mannose, SeNPs, M-SeNPs, C/S-Se and C/S-MSe all were highly biocompatible as indicated by IC₅₀ values within 20 µM in both HIEC and PMs, and cell viability above 70% (Fig. 2D-F & Figure S2B-D). NCM460 appeared to be more sensitive to the Na₂SeO₃ as indicated by cell viability that declined below 50% at 2.5 ~ 5 µM with a recovery to > 85% by M-SeNPs and SeNPs within 20 µM. Overall, these results showing the biocompatibility of M-SeNPs, which protect the IECs from DSS-induced injury.

Fig. 2.

M-SeNPs are localized to the lysosomes of IECs and protected cells from DSS induced damage. (A) The localization of C6-labeled M-SeNPs and SeNPs in lysosome of NCM460 cells were assessed by confocal microscopy. Nuclei were stained with DAPI (Blue). (B) M-SeNPs protected NCM460 cells against DSS-induced damage in a dose-dependent manner. The NCM460 cells were treated with different concentrations of M-SeNPs (0, 0.25, 1, 2, 4 µM) and 20 mg/mL of DSS with each group for 12 h. (C) Schematic illustrating the absorption profile of M-SeNPs in IECs. Cytotoxicity of M-SeNPs in (D) NCM460 cells, (E) HIECs and (F) PMs. Cells were treated with different Se concentrations of M-SeNPs, SeNPs and Mannose for 24 h, and cell viability assessed by CCK-8 assay. Mannose concentration is equal to the calculated loading level on M-SeNPs. Data are expressed as means ± SD (n = 3). *p < 0.05, ***p < 0.001

M-SeNPs suppress inflammation in IECs by modulating GPX and CD206 expression

Mannose is a natural bioactive monosaccharide that binds specifically to mannose receptors (CD206) on the cell surface and exhibits anti-inflammatory as well as anti-oxidative activities [15, 52]. CD206 has been established as a marker of M2 macrophages, but less is characterized regarding its function in IECs, which led us to investigate the effect of M-SeNPs against inflammation in IECs. Results from immunofluorescence assays showed the expression of CD206 on NCM460 was upregulated by M-SeNPs (Fig. 3A). And, compared with an equal dose of mannose, M-SeNPs showed a more prominent effect of promoting CD206 expression on NCM460 cells (Fig. 3B-D) that may be attributed to SeNPs reducing the effective dosage of loading ligand as demonstrated in our previous work [30]. The upregulation of CD206 provides a rationale for the targeted functionality of M-SeNPs, and we next evaluated the expression of GPX1, GPX2, GPX3 and GPX4 in NCM460 in response to M-SeNPs under LPS stimulation by conducting western blot and real-time PCR. Results showed that the inflammatory state in NCM460 led to downregulated GPX1, GPX2, GPX3 and GPX4 protein levels while M-SeNPs reversed this effect (Fig. 3B-C). In comparison, the effects of mannose or SeNPs were lower than M-SeNPs. Real-time PCR results showed that GPX1, GPX2, GPX3 and GPX4 mRNA were markedly decreased by LPS, with mannose and SeNPs all showing mitigating effects that were less than M-SeNPs (Fig. 3D). NCM460 cells were evaluated for LPS-induced inflammation, with expression of inflammatory cytokines in NCM460 analyzed by real-time PCR. Pro-inflammatory factor mRNAs such as mcp-1, cox-2, nf-κb, il-6 and tnf-α were significantly suppressed by M-SeNPs, and anti-inflammatory il-10 was upregulated (Fig. 3E). These data support an effect of M-SeNPs on inhibiting LPS-induced inflammation in IEC, along with an effect on increased CD206 and GPX expression.

Fig. 3.

M-SeNPs treatment increases the expression of CD206 and GPX while suppressing LPS-induced inflammatory responses in IECs. (A) Immunofluorescence staining of CD206 in NCM460 cells. The nuclei were stained with DAPI (Blue). (B) Western blotting and (C) densitometric analysis of CD206, GPX1/2, GPX3 and GPX4 relatively in NCM460 cells. (D) RT-PCR analysis of mRNA levels of CD206, GPX1, GPX2, GPX3 and GPX4 in NCM460 cells. (E) Heatmap of RT-PCR analysis of inflammatory factors in NCM460 cells. NCM460 cells were treated with 1 µM of M-SeNPs, SeNPs or mannose (mannose concentration is equal to the calculated loading level on M-SeNPs) for 18 h, respectively, followed by LPS (100 ng/mL) for 6–12 h for mRNA or protein detection, respectively. * represents the significance compared with the control. Data in (C, E) are presented as mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001

M-SeNPs ameliorate the severity of DSS-induced colitis in mice

We examined the efficacy of M-SeNPs in a model involving DSS-induced colitis in mice. C57BL/6 mice were treated with 3% DSS in drinking water for a duration of 7 days. On day 7, the mice were gavaged with C/S-Se, Mannose or C/S-MSe (Fig. 4A). To overcome physiologic barriers and deliver M-SeNPs to the inflamed colonic area, M-SeNPs were encapsulated in a colon-targeted hydrogel polymerized with CS and SA to form C/S-MSe [44, 53]. We used ICP-MS to examine the time course of Se biodistribution at 0, 2, 6, 12, 24, and 48 h in stomach, small intestine and colon, confirming effective delivery of the cargo to the colon and prolonged retention using the hydrogel system (Figure S3). Mice in the DSS group exhibited body weight loss, but C/S-MSe mitigated this weight loss (Fig. 4B). Furthermore, C/S-MSe significantly reduced the disease activity index (DAI) scores during progression of disease, surpassing the efficacy of both Mannose and C/S-Se groups (Fig. 4C). Compared with DSS group, splenomegaly was alleviated in both C/S-Se and C/S-MSe groups that generally correlated with decreased inflammation and anemia, and the effect was more pronounced in C/S-MSe treatment group (Fig. 4D-E). In addition, colon shortening caused by DSS was considerably reversed in C/S-MSe-treated groups (Fig. 4F-G). Hematoxylin and eosin (H&E)-staining in paraffin-embedded segments of colons further confirmed the potent therapeutic effect of C/S-MSe, with the histological disease scores recorded (Fig. 4H-I). Severe pathology that included extensive immune cell infiltration and significant disruption of epithelial structures throughout the proximal and distal colon was observed in DSS group. C/S-MSe markedly reduced immune cell infiltration and led to intact epithelial structures, resulting in notably decreased histological disease scores. Overall, these results show that M-SeNPs ameliorate the severity of DSS-induced colitis.

Fig. 4.

Therapeutic effects of M-SeNPs in the mouse model of DSS-induced colitis. (A) Schematic showing the experimental design for DSS-induced colitis. (B) The body weight of mice during 14 d in different treatment groups. All data were normalized relative to the day 0. (C) Daily disease activity index (DAI) scores in different treatment groups. (D) Digital images of spleens harvested from the different groups. (E) Spleen index calculated by the ratio of spleen to body weight. (F) Representative ex vivo images of colons. (G) Colon length in different groups of mice. (H) Representative histological images of proximal, middle and distal colon sections stained with H&E. (I) Histological disease score of different treatment groups. * represents the significance compared with the control. Data in (B, C, E, G) are represented as mean ± SD. *p < 0.05, **p < 0.01. Each group includes 9 independent mice

M-SeNPs diminish oxidative stress by facilitating colonic GPX expression in DSS-induced colitis

In the context of UC, oxidative stress plays a crucial role in the initiation, development, exacerbation and recurrence of the inflammatory disease within the gastrointestinal tract. To evaluate the impact of M-SeNPs on oxidative stress in DSS-induced colitis, we assessed the levels of MDA, GSH and the activity of GPX in colons, ranging from proximal to distant regions, as well as in various organs including liver, kidney, spleen and small intestine. As shown in Fig. 5A, established disease indicators such as oxidative stress and MDA were reduced in with C/S-MSe compared with DSS group. DSS was shown to induce systemic oxidative stress in all tissues, while treatment with C/S-MSe effectively mitigated this effect (Fig. 5B). GSH levels as well as the GPX enzymatic activity were found to be significantly decreased by DSS treatment in vivo, but this effect was reversed by oral administration of C/S-MSe (Fig. 5C-F). Se levels within colon, liver, kidney, spleen and small intestine were examined using ICP-MS, and Se predominantly accumulated in the middle and distal regions of the colon, and treatment with C/S-MSe restored the Se levels back to levels in the healthy control (Fig. 5G). Moreover, Se levels in other tissues affected by DSS exposure were also recovered by C/S-MSe administration (Fig. 5H). These results suggest that the administration of M-SeNPs embedded in CS/SA hydrogel could effectively mitigates oxidative stress levels in colitis, which correlated with increased Se.

Fig. 5.

M-SeNPs mitigate oxidative stress in DSS-induced colitis. MDA levels were determined in the (A) whole colon as well as in (B) liver, kidney, spleen and small intestine. GSH levels were investigated in the (C) whole colon, (D) liver, kidney, spleen and small intestine. GPX activity were also evaluated in the (E) whole colon and (F) several other tissues. Total Se concentration in the (G) colon, (H) liver, kidney, spleen and small intestine measured by ICP-MS. Data are represented as mean ± SD (n = 3)

Given that Se functions through its incorporation into selenoproteins and the GPXs are antioxidant enzymes, we investigated the expression of GPX1, GPX2, GPX3 and GPX4 in the colons of colitis mice. GPX proteins in the middle and distal region of colon exhibited a more pronounced change with DSS treatment (Fig. 6A-F), which corresponded to the results of colonic Se levels shown above. The DSS group exhibited a significant down-regulation of colonic GPX1, GPX2, GPX3 and GPX4 protein expression, but C/S-MSe supplementation reversed levels close to (or even higher than) the healthy control group. GPX1, GPX2, GPX3 and GPX4 mRNA levels showed a similar effect of C/S-MSe on reversing DSS-induced decreases (Fig. 6G). While C/S-MSe promoted the expression of CD206 (mannose receptor) in the middle and distal sections of the colon, DSS treatment also triggered the up-regulation of CD206 in this case (Fig. 6). This may be due to the fact that DSS is a glycan per se, which could also stimulate the up-regulation of the mannose receptor. Collectively, these data suggest that M-SeNPs exhibit antioxidant capacity in DSS-induced colitis by up-regulating GPXs.

Fig. 6.

M-SeNPs promote the expression of colonic CD206 and GPXs. Western blot analyses of the CD206, GPX1/2, GPX3, GPX4 in mice (A) proximal, (C) middle, (E) distal colon, and the consistent (B, D, F) densitometric quantification of these bands. (G) RT-PCR analysis of mRNA levels of CD206, GPX1, GPX2, GPX3 and GPX4 in the whole colon (n = 3). * represents the significance compared with the control. Data in (B, D, F) are represented as mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001

M-SeNPs block NF-κB activation and exert an anti-inflammatory effect in IECs

NF-κB has been reported to play important role for IECs participating in innate immune defense in the pathogenesis of UC [54–56]. To gain further insight into the underlying mechanism by which M-SeNPs impact IECs in UC, we investigated the NF-κB signaling pathway. NCM460 cells were primed with 20 mg/mL DSS for 24 h with or without M-SeNPs, (Fig. 7A&B). Nuclear p65 and cytoplasmic IK-Bα showed significantly higher phosphorylation by DSS treatment, while M-SeNPs treatment caused a significant reversal of this effect. In mouse colon tissues, immunofluorescence data showed a significant decrease of NF-κB in C/S-MSe group mice in IECs throughout the colon compared with DSS group (Fig. 7C-E). High colonic mRNA levels for IL-1β, IL-6, MCP-1, TNF-α, NF-κB and Cox-2 that marked inflammation in DSS group were down-regulated by C/S-MSe administration, while IL-10 was elevated (Fig. 7F). Altogether, the effects of M-SeNPs in restoring anti-inflammation by blocking NF-κB activation is consistent with its capacity to ameliorate the pathogenesis of colitis in vivo.

Fig. 7.

M-SeNPs alleviate colonic inflammation by inhibiting NF-κB activation in IECs. (A) Western blot analysis and (B) the quantification of NF-κB signal pathway in NCM460 cells stimulated with DSS. The cells were treated with 1 µM of M-SeNPs and DSS (20 µM) for 24 h. Immunofluorescence images of NF-κB in the (C) proximal, (D) middle and (E) distal colon tissue with various treatment. (F) The relative mRNA expression levels of inflammatory cytokines in whole colon (n = 3). Data in B are represented as mean ± SD, ***p < 0.001

Discussion

The innate immune function of IECs plays crucial role in maintaining intestinal homeostasis which has been consistently underestimated. IECs constitute the frontline in the intestinal encounter and recognition of pathogenic microorganisms. They express Pattern Recognition Receptors (PRRs) and function as dynamic signal transducers in the microbial environment, integrating signals into their own immune regulatory capabilities and actively participate in the directing of mucosal immune cell responses [6]. The sustained activation of the NF-kB pathway within IECs induced by PRRs triggers inflammation and tissue damage, disrupting intestinal immune homeostasis and exacerbating the progression of IBD [57]. Most existing studies focus more on colitis resolution strategies based on the restoration of the barrier function of IECs [43, 58]. In our study, the promising hydrogel microbeads (CS/SA)-embedding M-SeNPs demonstrated the ability to selectively target the colon and inhibit the activation of the NF-kB signaling pathway in IECs, mitigating inflammation and oxidative stress in mice model of DSS-induced colitis. This approach selectively modulates the intrinsic immune function of IECs using a nanoparticle-based therapeutic agent, offering an alternative method for treating IBD with the potential to resolve persistent inflammation.

Mannose, a naturally occurring sugar, exerts beneficial effects in IBD through its intrinsic anti-inflammatory properties [21, 22]. The binding of mannose and C-type lectin (CD206) assumes a pivotal role in maintaining intestinal homeostasis and regulating the inflammatory response associated with IBD. Our previous research demonstrated that a naturally mannose-rich oligosaccharide (MRO) can alleviate intestinal inflammation by regulating macrophages reprogramming [29]. However, to our knowledge, there is little existing research from the perspective of IECs that explores targeting regulating CD206 through specific binding of M-MRs to mitigate UC. In this study, we modified SeNPs with mannose, providing a specific and distinct target on IECs, differentiating this approach from previous studies, and underscoring its significance and necessity [29, 43, 58]. Indeed, the dysregulation of mannose metabolism and the excessive circulating mannose levels caused by disrupted IECs function are correlated with the occurrence of IBD. Zuo et al. indicated the therapeutic effect of mannose on IBD relies on high-dose exogenous supplementation [23]. However, mannose cannot endure within IECs, with the majority being rapidly excreted directly through urine within a short period [59]. The limited affinity of M-MRs (CD206) constrains the substantial accumulation of mannose in IECs, thus impeding its optimal functionality. In this study we modified D-mannose on SeNPs, in contrast to free mannose, providing more or less shape complementarity to the lectin oligosaccharide binding site and consequently increasing the affinity of M-MRs. In vitro studies in Fig. 2A showed that IECs had significantly higher uptake of M-SeNPs compared to SeNPs alone. Biodistribution data of Se in mice from Figure S3 also revealed more accumulation of C/S-MSeNPs in colonic tissues than C/S-Se. Furthermore, our results in Fig. 6 showed a significant upregulation of CD206 in the mouse colon by C/S-MSe. This elevation can further enhance the adhesion and accumulation of mannose within IECs, facilitating a more robust anti-inflammatory response. Notably, apart from the conventional role solely as a mere endocytic receptor facilitating the internalization of extracellular material for clearance and antigen presentation, CD206 has been established by recent research that actively contributes to shaping immune responses [60]. Although CD206 are over-expressed on colonic IECs in the context of IBD, this is not contradictory to the upregulation of CD206 induced by M-SeNPs and its associated anti-inflammatory effects [61]. Firstly, M-SeNPs can enhance uptake by increasing the clustering of CD206 on the surface of IECs [27, 28]. Secondly, CD206 is an endocytic receptor constantly cycling within cells, and the transient upregulation induced by M-SeNPs is eliminated through endocytosis, thereby not contributing to sustained reinforcement of pathogen recognition, uptake, and immune response. Further investigation is underway to delve deeper into this aspect, and elucidation will be provided in subsequent studies.

Se deficiency has been identified as a risk factor in exacerbating intestinal injury and promoting inflammatory responses in UC. This element exist as an essential trace nutrient for humans, co-translationally incorporated as the 21st amino acid, selenocysteine, into selenoproteins that play a crucial role in modulating pathways involved in inflammation. Experiments conducted with Se-deficient mice models demonstrated a reduction in the severity of the colonic inflammation upon Se supplementation [62], and its protective effect might be mediated by specific selenoproteins such as GPXs [36]. For instance, research indicates that mice deficient in both GPx1 and GPx2 spontaneously develop early-onset ileocolitis and intestinal cancer after 6 to 9 months [63, 64]. Additionally, Christopher S et al. suggested an immunomodulatory role for GPX3 that limits the development of colitis-associated carcinoma [65]. And GPX4 is a well-known antioxidant enzyme that has been extensively studied and reported to be associated with UC alleviating [66]. In a separate randomized and double-blind clinical trial, the supplementation of Se was found to contribute to the reduction of inflammation and disease activity in patients with UC [67]. However, the narrow nutritional range of Se intake limits its utilization. Among all forms of Se, elemental Se (0) in nanoparticles, i.e. SeNPs are less toxic and more bioavailable, rendering it potentially substitutable for traditional selenium supplements in the field of biomedicine [37–40]. In vivo, SeNPs possesses a heightened efficacy in synthesizing selenoproteins like GPXs [68]. According to the present study in Figs. 6 and 7 and C/S-MSe supplementation upregulated the expression of the GPXs family proteins (including GPX1, GPX2, GPX3 and GPX4) in inflamed-colonic tissues, thus exerting promise anti-inflammatory and antioxidant effect in vivo. Moreover, depending on their size and surface functionality, SeNPs could exhibit multifunctional effects in target delivery and controlled release by conjugating with various ligands. Our previous study highlighted the potential of MRO-functionalized SeNPs in macrophage reprogramming and set the stage for the current study [29, 30]. Herein, we conducted the modification of D-mannose on SeNPs and investigated its synergistic intervention on DSS-induced experimental colitis in mice. Based on our current findings, M-SeNPs exhibit lower toxicity in PMs and NCM460 cell lines compared to sodium selenite. Additionally, it effectively reduces IECs cell death caused by DSS, providing further evidence of the biocompatibility and safety of M-SeNPs. To better deliver towards colon with minimal digestive enzyme degradation and systemic absorption, optimized M-SeNPs was encapsulated into a colon-targeted hydrogel delivery system as previously reported [44]. This hydrogel comprises 2 polymers, CS and SA. SA is a water-soluble natural polysaccharide with an anionic charge, commonly utilized as a drug carrier due to its superior biocompatibility and biodegradability. CS, a cationic polymer derived from the deacetylation of chitin, also offers excellent biocompatibility, biodegradability, and significant bioadhesive properties. These attributes make the CS/SA hydrogel an ideal candidate for colon-targeted drug delivery systems. The CS/SA hydrogel matrix effectively prevents premature M-SeNPs release in the stomach, ensuring prolonged retention and controlled release in the colon, where the pH levels typically range 5 ~ 6. In the C/S-MSe treated mice, we observed a significantly higher s elenium retention level in colonic tissue compared to other parts of the gastrointestinal tract, such as the stomach and small intestine. This indicates that the release of M-SeNPs from the hydrogel occurs mostly in the colonic lumen, helping to enhance their effectiveness on the colonic IECs. Hence, C/S-MSe targeted induced an upregulation of GPXs in inflamed colonic IECs, and thus represent a safe therapeutic material for UC.

Recently, NPs-mediated drug delivery systems for the therapy of diverse diseases are of great interest in biomedical field due to targeted drug delivery, extended drug release, and higher drug stability to avoid rapid clearance at plaque areas. Several research has begun to focus on the polysaccharides functionalized SeNPs as a potential therapeutic approach for IBD [43, 58]. The present study advances the concept by incorporating M-SeNPs into a CS/SA hydrogel. The unique properties of C/S-MSe, including their economic preparation, simple manufacturing technology, and targeted inhibitory effects on the inflammatory response in colonic IECs, make them a compelling candidate for clinical translation, particularly in the context of UC therapy. The nanoscaling of mannose facilitates its breakthrough in clinical applications, leading to an upregulation of receptor (CD206) expression levels and an enhanced affinity for M-MRs. This enables targeted accumulation of M-SeNPs within lesioned IECs. Simultaneously, the combined effects of mannose and the intrinsic anti-inflammatory and antioxidative properties of SeNPs synergistically contribute to suppressing the occurrence and progression of UC. While the preliminary findings are encouraging, certain aspects merit attention for further improvement and exploration. The translation of M-SeNPs into clinical practice requires a meticulous examination of their pharmacokinetics, long-term safety profile, and potential off-target effects. What’s more, the exact mechanism remains to be further elucidated.

Conclusion

This study introduced a novel and promising approach to treat UC using M-SeNPs nanosystem. The M-SeNPs was prepared and embedded into CS/SA hydrogel microbeads, which can be delivered targeting the colon to alleviate DSS-induced colitis by modulating GPXs expressions in IECs, likely due to blocking NF-κB activation and thus inhibit inflammation and oxidation. The modification of mannose on the nanoparticles strengthened the biocompatibility of M-SeNPs in IECs, which also promoted its targeted anti-inflammation function. After engulfed by IECs, M-SeNPs notably up-regulated CD206 (mannose receptor) expression that may promote positive feedback for specific binding and targeting capacity, as well as enhancing the levels of GPXs. Additionally, M-SeNPs nanoparticles were shown to mitigate the symptoms of DSS-induced colitis in mice, possessing anti-oxidant, anti-inflammation faculty by promoting GPXs expression that inhibit NF-κB signaling pathway. In summary, the presented results underscores the efficacy of M-SeNPs in mitigating oxidative stress and inflammation in colitis, laying the foundation for future clinical applications. However, challenges in mechanism clarity, long-term stability, and in-depth safety must be addressed to advance M-SeNPs as a viable therapeutic option for IBD patients, this study established the rational designing strategy for M-SeNPs and illuminate the application of this nanoparticle for UC therapy.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

H.Y. : Methodology, Data curation, Funding acquisition, Investigation, Methodology, Validation, Formal analysis, Writing - original draft, Revise. Z.W. : Methodology, Software. L.L. : Methodology, Investigation, Validation. X.W. : Investigation, Validation. X.W. : Methodology, Data curation, Investigation, Software. S.G. : Writing - original draft. Z.D. : Investigation, Validation. Z.C. : Funding acquisition. Q.L. : Supervision. P.R.H. : Writing - review & editing. J.H. : Project administration, Resources, Supervision. F.L. : Writing - original draft, Project administration, Resources, Revise. Z.H. : Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing - review & editing.

Funding

This work was supported by National Natural Science Foundation of China (82270471, 82300958, 82300506, 81870323), China Postdoctoral Science Foundation (2023M731316), Natural Science Foundation of Guangdong (2022A1515010253, 2021A1515010104), the Major Project of Science and Technology of Guangzhou (202206010033), Doctoral Workstation Foundation of Guangdong Second Provincial General Hospital (2023BSGZ002), and the Science Foundation of Guangdong Second Provincial General Hospital (TJGC-2022013).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All animal care and experimental protocols were conducted under the approval of the Animal Care and Ethics Committee of Jinan University (20210725-04).

Consent for publication

All authors have approved the manuscript and agree with the submission.

Competing interests

The authors declare no competing interests.