MnO2 and roflumilast-loaded probiotic membrane vesicles mitigate experimental colitis by synergistically augmenting cAMP in macrophage

Abstract

Background: Ulcerative colitis (UC) is one chronic and relapsing inflammatory bowel disease. Macrophage has been reputed as one trigger for UC. Recently, phosphodiesterase 4 (PDE4) inhibitors, for instance roflumilast, have been regarded as one latent approach to modulating macrophage in UC treatment. Roflumilast can decelerate cyclic adenosine monophosphate (cAMP) degradation, which impedes TNF-α synthesis in macrophage. However, roflumilast is devoid of macrophage-target and consequently causes some unavoidable adverse reactions, which restrict the utilization in UC.

Results: Membrane vesicles (MVs) from probiotic Escherichia coli Nissle 1917 (EcN 1917) served as a drug delivery platform for targeting macrophage. As model drugs, roflumilast and MnO₂ were encapsulated in MVs (Rof&MnO₂@MVs). Roflumilast inhibited cAMP degradation via PDE4 deactivation and MnO₂ boosted cAMP generation by activating adenylate cyclase (AC). Compared with roflumilast, co-delivery of roflumilast and MnO₂ apparently produced more cAMP and less TNF-α in macrophage. Besides, Rof&MnO₂@MVs could ameliorate colitis in mouse model and regulate gut microbe such as mitigating pathogenic Escherichia-Shigella and elevating probiotic Akkermansia.

Conclusions: A probiotic-based nanoparticle was prepared for precise codelivery of roflumilast and MnO₂ into macrophage. This biomimetic nanoparticle could synergistically modulate cAMP in macrophage and ameliorate experimental colitis.

Keywords: Macrophage; Membrane vesicles; MnO2; Roflumilast; Ulcerative colitis; cAMP.

Scheme 1

The schematic procedure of various nanoparticles preparation and the potential mechanism of UC treatment. A Rof@MVs, MnO₂@MVs and Rof&MnO₂@MVs were synthesized by sonication. Gelatin NPs, Gelatin nanoparticles. EcN, Escherichia coli Nissle 1917. B After enema, Rof&MnO₂@MVs passed through the defective epithelial layer in the inflammatory mucosa and were swallowed by macrophage in the lamina propria. MnO₂ was reduced into Mn²⁺ by H₂O₂ and Mn²⁺ bonded with adenylate cyclase (AC). Besides, roflumilast inhibited the activity of PDE4. Then roflumilast and MnO₂ increased the concentration of cAMP, in terms of elimination and production respectively. As a second messenger, cAMP could bind with protein kinase A (PKA) to phosphorylate cAMP-response element binding protein (CREB). Phosphorylated CREB (p-CREB) entered the cell nucleus and downregulated TNF-α expression

Fig. 1

Characterization of MVs-based nanoparticles. A Dynamic light scattering diameter of MnO₂. B X-ray photoelectron spectroscopy of MnO₂. C The higher-resolution manganese spectra. D Dynamic light scattering diameter of Rof@MVs (Rof, 20 μg/mL). E Dynamic light scattering diameter of MnO₂@MVs (MnO₂, 80 μg/mL). F Dynamic light scattering diameter of Rof&MnO₂@MVs (Rof, 20 μg/mL; MnO₂, 80 μg/mL). G ζ potential of Rof@MVs (Rof, 20 μg/mL), MnO₂@MVs (MnO₂, 80 μg/mL) and Rof&MnO₂@MVs (Rof, 20 μg/mL; MnO₂, 80 μg/mL). H Scanning electron microscopy of Rof&MnO₂@MVs (white scale bar 100 nm, black scale bar 1 μm). I Stability of MVs-based nanoparticles in simulated colon fluid (SCF, pH 7.2–7.8) at 37 °C

Fig. 2

Macrophage-target of MVs-based nanoparticles. A Comparison of nanoparticle uptake in CT26 (blue) and RAW264.7 (orange) via flow cytometer. B Quantification of uptake ratio of the same nanoparticle between CT26 and RAW264.7 cells with signal intensity exceeding 100 were concluded (n = 5). C The uptake of various MVs-based nanoparticles in RAW264.7 via flow cytometer. D MVs could prompt the cargo MnO₂ to accumulate in RAW264.7. E Quantification of uptake ratio of different nanoparticles in RAW264.7 (n = 5). F Quantification of uptake ratio of various nanoparticles in CT26 (n = 5). G Analysis of MnO₂ uptake in RAW264.7 with or without MVs (n = 5). H Mean fluorescent intensity of FITC-labelled nanoparticles in RAW264.7 when observed under CLSM. (n = 3). I Representative fluorescent graph of FITC-labelled nanoparticles in RAW264.7 (scale bar 20 μm). These data were manifested as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns (none significance). Data was statistically analyzed via two-tail Student’s t test or one-way ANOVA multiple comparisons tests (Tukey’s test was used for comparison of multiple groups)

Fig. 3

The underlying biological mechanism of roflumilast and MnO₂. A The intracellular decomposition of MnO₂ by H₂O₂, illustrated via flow cytometer. B Quantifying ROS relative ratio of different nanoparticles in RAW264.7. Cells with signal intensity surpassing 1000 were analyzed (n = 3). C The effect of diverse formulations of Mn on the cytosolic cAMP. MnO₂@MVs-H means high concentration (5 μg/mL MnO₂). Similarly, MnO₂@MVs-L represents low concentration (0.5 μg/mL MnO₂). The Mn concentration of both MnO₂ and MnCl₂ was 5 μg/mL. To obtain the relative ratio, the concentrations of cAMP in groups were divided by the control (n = 4). D The effect of MVs-based nanoparticles on the cytosolic cAMP. (n = 3). E The phosphorylation of CREB in various MVs-based nanoparticles in Western Blot. F The relative quantification of p-CREB in Western Blot (n = 5). G The synergistical efficacy of roflumilast and MnO₂ in inhibiting TNF-α secretion from RAW264.7 (n = 3). Rof@MVs (Rof, 1.25 μg/mL), MnO₂@MVs (MnO₂, 5 μg/mL), Rof&MnO₂@MVs (Rof, 1.25 μg/mL; MnO₂, 5 μg/mL) and MVs (62.5 μg/mL). The concentration ratio of roflumilast and MnO₂ is 1:4. These data were manifested as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns (none significance). Data was statistically analyzed via one-way ANOVA multiple comparisons tests (Tukey’s test was used for comparison of multiple groups)

Fig. 4

The biodistribution of nanoparticles in colon after enema. A Experiment scheme. 3% (wt %) DSS was used to establish colitis via free drinking from day 1 to day 6. Mice were given enema with FITC-labelled nanoparticles and analyzed on day 6. B Representative fluorescent imaging of nanoparticles uptake in macrophage in Vivo (scale bar 20 μm). C The detention time of FITC-labelled nanoparticles in colon after enema via in vivo imaging system. Abbreviations: Rof, Rof@MVs-FITC; MnO₂, MnO₂@ MVs-FITC; Rof&MnO₂, Rof&MnO₂@ MVs-FITC

Fig. 5

The efficacy of MVs-based nanoparticles in ameliorating DSS-induced colitis. A The scheme of experiment. Mice were firstly raised 7 days for acclimation before colitis establishment. Then, 3% (wt %) DSS was used to establish colitis via free drinking from day 1 to day 6. During this process, enema was performed on day 2, 4 and 6 with Rof (Rof@MVs, roflumilast 1 mg/kg), MnO₂ (MnO₂@MVs MnO₂, 4 mg/kg), Rof&MnO₂ (Rof&MnO₂@MVs, roflumilast, 1 mg/kg and MnO₂, 4 mg/kg) and 5-ASA (1.25 mg/kg). From day 7, DSS was substituted with drinking water until euthanasia. On day 9, mice were sacrificed and colon was collected for further analysis. B Dynamic body weight mass in different groups during the experiment (n = 5). C Measurement of colon length in various groups (n = 5). D Analysis of intestinal permeability with FD4 (n = 5). E TNF-α level in colon tissue (n = 5). F Photograph of colon from mice (n = 5). The grid behind the specimen is 1 cm × 1 cm. G Representative images of H&E stain of colon in each group (scale bar 100 μm). H Representative graphs of goblet cells in each group by Alcian blue staining (scale bar 100 μm). I Immunofluorescence examination in Occludin of colon from each group (scale bar 20 μm). These data were manifested as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Data was statistically analyzed via two-way or one-way ANOVA multiple comparisons tests (Tukey’s test was used for comparison of multiple groups)

Fig. 6

The efficacy of MVs-based nanoparticles in gut microbes. A The Chao1 of different groups (n = 5). These data were manifested as mean ± SD. B The observed operation taxonomy units (OTUs) of different groups (n = 5). These data were manifested as mean ± SD. C The Simpson diversity index of various groups (n = 5). These data were manifested as mean ± SD. D Principal components analysis (PCA) of intestinal microorganism. E Representative top 30 bacteria in genus level. F The relative abundance of Escherichia–Shigella in genus (n = 5). G The relative abundance of Akkermansia in genus (n = 5). Rof: Rof@MVs; MnO₂: MnO₂@MVs; Rof&MnO₂: Rof&MnO₂@MVs. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Data in A, B was statistically analyzed via one-way ANOVA multiple comparisons tests (Tukey’s test was used for comparison of multiple groups). Data in C, F, G was statistically analyzed via nonparametric tests (Mann–Whitney test was used for comparison of two groups)