Characterization of PELNs
The PELNs were isolated from POL, and which abundantly accumulated at the 8/30% interface (band 1), and 30/45% interface (bang 2) of the sucrose gradient (Figure S1a). The size distribution of PELNs was characterized as exosome-like nanoparticles, and confirmed by electron microscopic examination (Figure S1b). Moreover, the hydrodynamic particle sizes of PELNs ranged from 30 to 400 nm, with an average size of 180 nm (Figure S1c). In addition, Zeta potential measurements demonstrated that PELNs had a negative zeta potential of -31.4 mV (Figure S1d). The lipidomic analysis showed that PELNs mainly contained digalactosyldiacylglycerol (DGDG, 24.47%), triglyceride (TG, 17.09%), and phosphatidylcholine (PC 9.28%) (Figure S1e).
To examine the stability of PELNs in the GIT, PELNs were incubated in different aqueous solutions that mimicked a stomach-like solution and a small-intestine-like solution, and then analyzed changes in their zeta potential and size. Their zeta potential had a reduction in negative charge in stomach-like solution, whereas which was weakly positively changed in small-intestine-like solution (Figure S1f). The results showed that the heterogeneity in their size was increased in both a stomach-like and small-intestine-like solution comparing with that incubated in the water for 2 h (Figure S1g). Strikingly, these results indicated that PELNs can maintain integrity, and resist to digestion during their passage through the GIT.
Oral Administration Of Pelns Protects Mice Against Dss-induced Colitis
To investigate their anti-inflammatory effects of PELNs, mice were randomly divided into four groups: healthy control group, DSS control group, PELNs-H group, and PELNs-L group abovementioned. Protocol was available in Fig. 1a investigating the potential of PELNs to alleviate UC. The body weight was gradually increased in healthy control group, whereas which was dramatically decreased in DSS control group during the entire experiment (Fig. 1b). Nevertheless, PELNs had the therapeutic potential against weight loss (Fig. 1b). Treatment with PELNs can significantly prevent the body weight loss compared to DSS control group (Fig. 1b). Moreover, PELNs-H seemed to be more effective in the prevention of the body weight loss than PELNs-L (Fig. 1b). Consistently, PELNs exerted an important role in the reduction of disease activity index (DAI) (Fig. 1c). The PELNs groups have significantly lower DAI than that in the DSS control group (Fig. 1c). Furthermore, the DAI in the PELNs group is comparable with that in the healthy control group (Fig. 1c). As expected, PELNs remarkably reduced a reduction of colon shortening (Fig. 1d and 1e). PELNs had significantly longer colon comparing to the DSS control group (Fig. 1d and 1e). Furthermore, the colon length of the PELNs-H group was slightly longer than that in the PELNs-L group, the fecal condition and mucosa of the colon in PELNs-H group was better than that in PELNs-L group, whereas which was appreciably shorter than that in the healthy control group (Fig. 1d and 1e). In addition, hematoxylin-eosin staining (H&E) and histological scores also confirmed that PELNs decreased accumulation of immune cells and damage of colonic epithelial barrier (Fig. 1g and 1f). Interestingly, the histological scores were significant lower in the PELNs-H group than that in the PELNs-L group (Fig. 1g). Therefore, PELNs-H have greater anti-inflammatory therapeutic effect compared with PELNs-L in the treatment of DSS-induced mice colitis.
As known, the levels of pro-inflammatory cytokines play a crucial role in intestinal inflammation. Interestingly, pro-inflammatory cytokines (IL-6, IL-12, IL-1β, and TNF-α) were dramatically elevated in DSS control group compared with the healthy control group in colon samples from C57 mice colitis using qRT-PCR (Fig. 2a, 2b, 2c, 2d). However, treatment of PELNs remarkably inhibited the expression of these pro-inflammatory cytokines (Fig. 2a, 2b, 2c, 2d). On the contrary, IL-10, as an anti-inflammatory, was considerably decreased in DSS control group compared with the healthy control group, whereas which was sharply increased in the PLENs administered group (Fig. 2e). As expected, an ELISA assay demonstrated that the secretion amount of these pro-inflammatory cytokines exhibited a remarkable increase in the DSS control group compared with the healthy control group in blood samples from C57 mice colitis (Fig. 2f, 2g, 2h, 2i). However, the levels of these pro-inflammatory cytokine (IL-6, IL-12, IL-1β, and TNF-α) secretion were significantly reduced in the PELNs group (Fig. 2f, 2g, 2h, 2i). Accordingly, the secretion level of IL-10 was detected, and revealed that PLENs promoted the secretion profiles of IL-10 (Fig. 2j). Moreover, PELNs-H have more excellent performance in the decrease of pro-inflammatory cytokines and increase of anti-inflammatory cytokines than PELNs-L.
Oral Administration Of Pelns Protects Phlpp2 Mice Spontaneous Colitis
To further evaluate the anti-inflammatory effects of PELNs in a chronic colitis model, PHLPP2−/− mice were chosen due to we previously demonstrated that PHLPP2−/− mice spontaneously developed colitis with a time-dependent within 4 weeks post weaning. Subsequently, PHLPP2−/− mice were randomly divided into 3 groups: healthy control group, DSS control group and PELNs-H group. Surprisingly, PELNs-H treated mice failed to develop spontaneous colitis, owing to which can effectively prevent the body weight loss, colon shortening, immune cells infiltration, and histological scores compared with PHLPP2−/− mice not treated PELNs-H (Fig. 3).
Strikingly, qRT-PCR and ELISA results showed that oral administration of PELNs-H effectively suppressed the expressions of pro-inflammatory cytokines (IL-6, IL-12, IL-1β, and TNF-α), in colon tissues and blood samples of PHLPP2−/− mice spontaneous colitis. However, PLENs prominently increased the expression of IL-10 (Figure S2).
In vivo distribution of PELNs
The accumulation of oral nanotherapeutics in the desired colon is critical to exert therapeutic effectiveness against UC. To determine the PELNs biodistribution after oral administration, mice with UC were gavaged with IRDye 800CW -labeled PELNs (100 mg/g) for 3, 6, 12 and 24h. Near-infrared imaging revealed that IRDye 800CW-labeled PELNs was visible in the colon at the 3 h, and gradually increased at the 6 h, whereas fluorescent signals detected was steadily decreased in the colon at the 24 h (Fig. 4). However, no conspicuous fluorescence signals were detected in the colon when PELNs administered the healthy mice (Fig. 4). Subsequently, the accumulation of fluorescent signals in the heart, liver, spleen, lung, and kidney was examined, and found that there were very few fluorescent signals detected in these vital organs when PELNs treated mice colitis at 3, 6, 12 and 24h (Fig. 4). Collectively, PELNs can specially target to the inflamed colon in the treatment of mice colitis.
Pelns Distinctly Altered The Diversity Of The Intestinal Microflora
The intestinal microbiota profiles were analyzed by 16S rDNA sequencing. The abundance of gut microbiota could be reflected by calculating the number of OTUs and further be observed the richness and evenness through different α-diversity indexes. As shown in Fig. 5a, the OTUs of colitis mice was only 590 and was up-regulated significantly by PELNs. Further analyzed the effect of PELNs on α-diversity through five indicators, including ACE (Fig. 5b), Chao 1 (Fig. 5c), PD-tree (Fig. 5d), Simpson (Fig. 5e) and Shannon (Fig. 5f), and found that DSS induced significant decrease of α-diversity (ACE, P = 0.00002; Chao 1, P = 0.000016; PD-tree, P = 0.000171; Simpson, P = 0.000257; and Shannon, P = 0.000015), while PELNs could significantly up-regulate (ACE, P = 0.000775; Chao 1, P = 0.000707; PD-tree, P = 0.001010; Simpson, P = 0.107145; and Shannon, P = 0.003565). β-diversity usually used to judge the difference between groups of gut microbiota, the β-diversities of the intestinal microbiota across both cohorts were assessed using PCA (Fig. 5g), PCoA (Fig. 5h), and NMDS (Fig. 5i), consistent results were obtained with NMDS, PCA and PCoA (PCA1 + PCA2 = 79.44 > 50%, PCoA1 + PCoA2 = 52.40 > 50%, and stress = 0.078 < 0.1,) which showed that PELNs significantly change the microbial structure of colitis mice and approach to the healthy group.
Pelns Improve Dss Induced Changes In Dominant Microbiota And Mediated Significant Changes In Microbial Structure
As shown in Fig. 6a, the relative abundance of Firmicutes (51.26% vs. 19.36%, P = 0.000141), Bacteroidetes(20.70% vs. 16.91%, P = 0.374828), Patescibacteria (3.92% vs. 0.41%, P = 0.006606), Tenericutes (0.92% vs. 0.10%, P = 0.089697), Actinobacteria (0.46% vs. 0.13%, P = 0.071755), Lachnospiraceae (24.75% vs. 5.53%, P = 0.000056), Ruminococcaceae (10.81% vs. 3.18%, P = 0.001079), Muribaculaceae (12.08% vs. 5.09%, P = 0.008093), Lactobacillaceae (8.23% vs. 1.81%, P = 0.002318), Moraxellaceae (14.34% vs. 8.82%, P = 0.502718), Erysipelotrichaceae (3.64% vs. 1.32%, P = 0.004203), Lachnospiraceae_NK4A136_group (15.23% vs. 2.53%, P = 0.000263), Lactobacillus (8.22% vs. 1.81%, P = 0.002315), Acinetobacter (14.34% vs. 8.82%, P = 0.502718), Ruminococcaceae_UCG-014 (4.30% vs. 0.51%, P = 0.001301), Candidatus_Saccharimonas (3.92% vs. 0.41%, P = 0.006644), Alistipes (2.61% vs. 2.02%, P = 0.514264), Acinetobacter_calcoaceticus (14.11% vs. 8.43%, P = 0.484667), Lactobacillus_gasseri (3.95% vs. 0.40%, P = 0.027309), Lactobacillus_murinus (2.80% vs. 1.30%, P = 0.177553), Lactobacillus_reuteri (1.12% vs. 0.03%, P = 0.000201), Bacteroides_caecimuris (1.82% vs. 1.20%, P = 0.531045), Helicobacter_sp_MIT_03-1616 (0.20% vs. 0.18%, P = 0.759178), were notably elevated in mice treated with PELNs in comparison to colitis group, while that of Proteobacteria (18.79% vs. 49.81%, P = 0.004250), Deferribacteres (2.28% vs. 9.67%, P = 0.010716), Epsilonbacteraeota (0.76% vs. 2.15%, P = 0.188453), Verrucomicrobia (0.10% vs. 0.42%, P = 0.221802), Cyanobacteria (0.05% vs. 0.09%, P = 0.253980), Enterobacteriaceae (1.37% vs. 39.56%, P < 0.0001), Deferribacteraceae (2.28% vs. 9.67%, P = 0.010716), Bacteroidaceae (2.31% vs. 7.14%, P = 0.070016), Rikenellaceae (3.19% vs. 3.83%, P = 0.631456), Escherichia-Shigella (1.27% vs. 39.13%, P < 0.0001), Mucispirillum (2.28% vs. 9.67%, P0.010716=), Bacteroides (2.31% vs. 7.14%, P = 0.070016), Staphylococcus (0.17% vs. 3.67%, P = 0.164005), Bacteroides_thetaiotaomicron (0.31% vs. 5.87%, P = 0.029962), Staphylococcus_sciuri (0.12% vs. 3.61%, P = 0.162046), Mucispirillum_sp_69 (0.75% vs. 3.15%, P = 0.011562), Enterococcus_faecalis (0.13% vs. 2.17%, P = 0.037956) were decreased.
Welch's t test was used to analyze the microbial changes induced by PELNs. There was a higher abundance of the phyla (Fig. 6b) Firmicutes and Patescibacteria, while decreased abundance of Proteobacteria and Deferribacteres. The relative proportion of the family (Fig. 6c) Lachnospiraceae, Ruminococcaceae, Muribaculaceae, Lactobacillaceae, Saccharimonadaceae, Erysipelotrichaceae, Erysipelotrichaceae, Prevotellaceae, Clostridiales_vadinBB60_group, Burkholderiaceae, Vibrionaceae, Eggerthellaceae, Eggerthellaceae, Peptococcaceae, and Bacillaceae increased remarkably, and that of Enterobacteriaceae, Deferribacteraceae, Enterococcaceae, and Tannerellaceae decreased compared to the colitis group. Furthermore, the abundant genera (Fig. 6d) relative to the PELN group were Lachnospiraceae_NK4A136_group, Lactobacillus, Ruminococcaceae_UCG-014, Candidatus_Saccharimonas, Ruminiclostridium_6, Dubosiella, Parasutterella, Alloprevotella, GCA-900066575, Ileibacterium, Photobacterium, Turicibacter, Lachnospiraceae_UCG-006, Ruminococcus_1, Bilophila, Enterorhabdus, Butyricicoccus, ASF356, UBA1819, Harryflintia, GCA-900066225, Muribaculum, and Acetatifactor, whereas Escherichia-Shigella, Mucispirillum, Enterococcus, Rikenellaceae_RC9_gut_group, and Parabacteroides had low abundance. At the species level (Fig. 6e), PELNs drastically elevated the relative proportion of Lactobacillus_gasseri, Lactobacillus_reuteri, Alistipes_inops, Ileibacterium_valens, Photobacterium_damselae_subsp_damselae, and Ruminococcus_flavefacien, and lowered that of Bacteroides_thetaiotaomicron, Mucispirillum_sp_69, Enterococcus_faecalis, and Parabacteroides_goldsteinii. (The Welch's t test results of Control group vs. DSS group are provided as supplementary materials Figure S3).
PELNs exert therapeutic effect of mice colitis by inducing the differentiation of DP CD4 + CD8 + T cells
As reported, DP CD4+CD8+ T cells can maintain a steady-state of the gut mucosa, and inhibit pro-inflammatory cytokine release during pathogenic infection36. Moreover, we previously demonstrated that the population of DP CD4+CD8+ T cells was significantly decreased in UC mice colon samples35. Consequently, we investigated whether the oral administration of PELNs have an effect in the differentiation of DP CD4+CD8+ T cells in mice colitis. Strikingly, the population of DP CD4+CD8+T cells was significantly increased in the PELNs group compared to the DSS control group using FCM (1.1% vs. 0.75%, P < 0.01, Fig. 7a). Nevertheless, there was significant difference of the population of DP CD4+CD8+T cells was detected between the PELNs group and the healthy group (1.1% vs. 2.45%, P < 0.05, Fig. 7a). Similarly, IF revealed that the population of DP CD4+CD8+T cells was dramatically increased in the PELNs group compared to the DSS control group (Fig. 7b). Furthermore, the population of DP CD4+CD8+T cells was comparable among the PELNs group and the healthy group (Fig. 7b).
Biosafety Of Orally Administered Pelns
To evaluated the biosafety of orally administered PELNs, the blood serum samples and vital organs (heart, liver, spleen, lung, and kidney) were obtained, when mice were daily given 50mg/g and 100mg/g PELNs for 5 days. H&E staining for histological analysis of the vital organs found no obvious abnormalities of organs damage in PELNs-administered group (Fig. 8a). H&E staining of the vital organs in the healthy control group, the DSS group, PELNs-L group and PELNs-H group was shown in Figure S4. Moreover, blood cell count, cardiac enzymes, glutamic pyruvic transaminase (ALT), aspartate aminotransferase (AST), Serum Creatinine (CREA), and Serum Urea (UREA) did not significantly change between the normal control group, DSS group and PELNs-administered group (Fig. 8b, 8c, 8d, 8e).