Bacterial depletion ameliorates pancreatic injury and the inflammatory response in SAP
To investigate whether the gut microbiota is involved in the progression of SAP, ATBx treatment was used to deplete the gut microbiota of mice (Supplementary Fig. 1). Firstly, the WT mice were divided into four groups: the blank control (BC), ATBx, SAP, and ATBx-SAP groups (Fig. 1a). Compared to SAP group, hematoxylin and eosin (H&E) staining revealed that inflammation and necrosis of the pancreas were significantly decreased in ATBx-SAP group (Fig. 1b, c), and serum amylase levels were decreased in ATBx-SAP group (Fig. 1d). These results suggest that bacterial depletion in gut ameliorates pancreatic injury of SAP.
Then all 23 inflammatory mediators were increased in SAP, but the levels of IL-6, IL-17A, IFN-γ, TNF-α, and IL-4 were significantly reduced after ATBx treatment (Fig. 1e). In addition, we previously showed that high triglyceride (TG) levels affect pancreatic injury in pancreatitis20. As shown in Fig. 1f, serum TG and total cholesterol (TC) were higher in SAP, and ATBx treatment significantly reduced serum TG concentration. Our data suggest that bacterial depletion in the gut can ameliorate inflammatory response and metabolic abnormalities of SAP.
Bacterial depletion reverses SAP-induced gut microbial metabolism imbalance
To explore the role of gut microbial metabolism in SAP, the gut microbiota composition and indices of microbial metabolism were analyzed. Across the BC, ATBx, and SAP groups, PCoA analysis revealed a distinct microbial composition among them (Fig. 2a). Taking the BC group as a reference, the top 10 taxonomic phyla and families separately in terms of abundance were presented with the proportion of changes in each group (Fig. 2b). At the phylum level, these was a decrease in the abundances of Firmicutes and Desulfobacterota and an increase in the abundances of Proteobacteria and Verrucomicrobia in SAP compared to BC group (LDA score > 3.5) (Fig. 2c). At the family level, these was a decrease in the abundances of Lachnospiraceae, Erysipelotrichaceae, and Desulfovibrionaceae and an increase in the abundances of Enterobacteriaceae, Akkermansiaceae, and Lactobacillaceae in SAP compared to BC group (Fig. 2c). The abundances of Proteobacteria and Enterobacteriaceae at the phylum and family levels were significantly increased after ATBx treatment, accounting for over 80% in ATBx group and over 60% in ATBx-SAP group (Fig. 2b, d and Supplementary Fig. 2). These data suggest that there is significant gut microbiota disorders in SAP and that the gut microbiota is indeed largely depleted after ATBx treatment.
Next, to investigate the impact of SAP-induced gut microbiota disorders on microbial metabolism, metabolomics analysis was used to evaluate changes in the metabolic components of each group. PCA analysis showed a clear separation of the metabolic components among the BC, ATBX, and SAP groups (Fig. 2e). For metabolites with significant changes in abundance in the SAP vs. BC group (fold change ≥ 2 or ≤ 0.5 and p ≤ 0.001), we showed the distribution of the top 10 metabolic classes. In the SAP vs. BC group, increased microbial metabolites were dominated by lipids metabolism, with 29.82% and 17.54% of fatty acyls and glycerophospholipids, respectively. And decreased microbial metabolites were dominated by carbohydrate and amino acid metabolism, with 33.33% and 8.51% of carboxylic acids and derivatives and organooxygen compounds, respectively (Fig. 2f). Compared to BC or SAP group, ATBx treatment exhibited opposite differences in the abundances of significantly altered gut microbial metabolites (Fig. 2g and Supplementary Fig. 3a, b). These data suggest that gut microbial metabolism is similarly imbalanced along with gut microbiota disorders in SAP.
Indeed, according to the ecological distribution of overall abundance of microbial metabolism in BC mice, the overall abundance of fatty acyls & glycerophospholipids was essentially equal to that of carboxylic acids and derivatives & organooxygen compounds (Fig. 2h). ATBx treatment decreased the overall abundance of microbial metabolism, specifically an increase in fatty acyls & glycerophospholipids and a decrease in carboxylic acids and derivatives & organooxygen compounds (Supplementary Fig. 3c-e). To more directly analyze the effect of ATBx treatment on SAP-induced microbial metabolism imbalance, the metabolites in the intersection of ATBx-SAP vs. SAP group and SAP vs. BC group were analyzed (Fig. 2i). ChemRICH class enrichment showed that SAP-induced changes in microbial metabolism were the opposite of those observed following ATBx treatment (Fig. 2j). Specifically, in carboxylic acids and derivatives & organooxygen compounds, the metabolites showing significantly increased abundances were mainly oligosaccharides and peptides after depleting the gut microbiota, with the significant changes in trehalose (Fig. 2k). This indicates that bacterial depletion reverses SAP-induced microbial metabolism imbalances (Fig. 2l). Based on these data, we hypothesized that trehalose in oligosaccharides is involved in the progression of SAP.
Trehalose ameliorates the severity of severe acute pancreatitis
Trehalose has been shown to play a beneficial role in the treatment of various diseases21. Combined with the previous results, we further determined whether trehalose has a beneficial role in SAP. WT mice were randomized into the BC, trehalose (Tre), SAP, and Tre-SAP groups, and a graded dose of trehalose was orally administered daily for 15 days (Fig. 3a). The possible effects of trehalose was excluded by adding a 0.2% trehalose group. Compared to the SAP group, pancreatic inflammation, and necrosis were significantly alleviated in the Tre-SAP group with 0.2%, 0.5%, 1%, or 2% trehalose (Fig. 3b, c). Oral 5% trehalose did not change the pathology scores of SAP, and we found that 5% trehalose can cause diarrhea and increased urine output in SAP. Serum amylase levels were not altered after trehalose treatment (Fig. 3d). These data confirm that trehalose can ameliorate pancreatic injury in SAP.
Next, according to the inflammatory cytokines profile, 0.2% trehalose significantly reduced the expression of TNF-α, MCP-1, IL-6, and IL-17A in SAP, and MCP-1 and TNF-α levels were significantly reduced in the Tre group compared to BC group (Fig. 3e-g). In addition, trehalose treatment effectively reduced SAP-induced increases in TG levels but not TC levels (Fig. 3h). These findings suggest that trehalose treatment ameliorates inflammatory response and metabolic abnormalities of SAP.
Trehalose modulates the gut microbial metabolism homeostasis in SAP
Considering the significant change in trehalose observed in the ATBx-SAP group, we further explored whether trehalose could act on the gut microbiota and microbial metabolism. Differences in the gut microbiota and microbial metabolism were analyzed among the BC, 0.2% Tre, SAP, and 0.2% Tre-SAP groups. PCoA analysis showed that the gut microbiota among the BC, 0.2% Tre, and SAP groups were entirely separated (Fig. 4a). First, the gut microbiota disorders of SAP in the phyla and families were consistent with previous descriptions (Supplementary Fig. 4). At the phylum level, there was an increase in the abundances of Bacteroidota and Proteobacteria and a decrease in the abundances of Firmicutes and Deferribacterota in the Tre group compared to BC group, and trehalose significantly promoted the abundance of Bacteroidota and suppressed Firmicutes in SAP group (Fig. 4b-d, Supplementary Fig. 4a, b). At the family level, the abundance of Muribaculaceae increased, whereas that of Lactobacillaceae decreased in the Tre or Tre-SAP group (Fig. 4b-d, Supplementary Fig. 4c, d). Taken together, these findings confirm that oral trehalose can modulate gut microbiota disorders, specifically the expansion of Muribaculaceae and inhibition of Lactobacillaceae.
Next, PCA analysis revealed a small difference between SAP and Tre-SAP groups (Fig. 4e). Compared to BC group, the increased metabolites in SAP group were predominantly fatty acyls & glycerophospholipids, and the decreased were predominantly carboxylic acids and derivatives & organooxygen compounds, which is consistent with previous data (Fig. 4f). According to the previous screening criteria (fold change ≥ 2 or ≤ 0.5 and p ≤ 0.001), there were few significantly different metabolites in Tre-SAP vs. SAP group (Fig. 4g and Supplementary Fig. 5a). These suggested that trehalose treatment did not cause dramatic significant changes in gut microbial metabolism as ATBx treatment. Therefore, we choose another commonly used analytical model, OPLS-DA model (Supplementary Fig. 5b). Trehalose treatment significantly altered the abundances of 74 metabolites in the intersection of Tre-SAP vs. SAP group and Tre vs. BC group (VIP ≥ 1) (Supplementary Fig. 5c). Carboxylic acids and derivatives & organooxygen compounds were the main classes in increased metabolites, while carboxylic acids and derivatives & fatty acyls were the main classes in decreased metabolites (Supplementary Fig. 5d). Moreover, the abundances of 93 metabolites were significantly altered in the intersection of Tre-SAP vs. SAP group and SAP vs. BC group (VIP ≥ 1) (Fig. 4h). Trehalose treatment increased the abundances of organooxygen compounds inhibited in SAP group, and decreased the abundances of fatty acyls elevated in SAP group (Fig. 4i). ChemRICH class enrichment also showed that trehalose treatment resulted in decreased abundance of fatty acyls and increased abundance of organooxygen compounds in Tre-SAP group compared to SAP group (Fig. 4j-k). These results indicate that trehalose ameliorates SAP by modulating gut microbial metabolism, specifically increasing carbohydrate metabolism and decreasing lipids metabolism.
Furthermore, for significantly different metabolites after trehalose treatment, pathway enrichment analysis revealed that the pathways significantly associated with trehalose were galactose metabolism, pentose phosphate pathway, and biosynthesis of unsaturated fatty acids (p < 0.01) (Fig. 4m). The metabolites related to trehalose were D-(+)-galactose and inositol in galactose metabolism, which were increased by trehalose in the Tre-SAP vs. SAP group compared to SAP vs. BC group (Fig. 4n). In the biosynthesis of unsaturated fatty acids, 16-hydroxyhexadecanoic acid, arachidonic acid and 8Z,11Z,14Z-eicosatrienoic acid were decreased by trehalose (Fig. 4n). Similarly, D-(+)-galactose was increased in the intersection of Tre-SAP vs. SAP group and SAP vs. BC group (Supplementary Fig. 5e). Taken together, these findings suggest that galactose metabolism is a key factor in the trehalose-mediated “microbial metabolism-gut-pancreatic axis”.
Trehalose ameliorates SAP in a gut microbiota-dependent manner
To further verify whether the role of trehalose depends on gut microbiota, we conducted FMT from BC or Tre mice to GF mice and labeled them "BC-FMT" or "Tre-FMT" group, respectively (Fig. 5a). First, in GF mice, oral trehalose did not ameliorate SAP but instead increased serum keratinocyte-derived chemokine (KC) and IL-10 levels (Fig. 5b-e). After FMT, pancreatic inflammation, and necrosis were significantly increased in the BC-FMT-SAP compared to the SAP group, and their pathological scores in the Tre-FMT-SAP group were lower than those in the BC-FMT-SAP group (Fig. 5b, c). Consistent with oral trehalose, trehalose-remodelled microbiota failed to reduce serum amylase levels in Tre-FMT-SAP group (Fig. 5d). Moreover, serum TNF-α, IL-6 and IL-17A were decreased in the Tre-FMT-SAP group compared to BC-FMT-SAP group (Fig. 5f). The serum TG level increased after FMT in GF mice, and the trehalose-remodelled microbiota modulated this change (Fig. 5g). These data confirm that the role of trehalose in SAP depends on gut microbiota.
Galactose metabolism and Muribaculaceae are key factors in trehalose-induced “microbial metabolism-gut-pancreatic axis”
To further understand the effect of trehalose on the “microbial metabolism-gut-pancreatic axis” in SAP, gut microbiota and microbial metabolism were analyzed after FMT in GF mice. There were some differences in the gut microbiota between BC-FMT-SAP and Tre-FMT-SAP groups (Fig. 6a). From the top 10 phyla (families) in terms of abundance, as in WT mice, bacteroidota (Muribaculaceae), Verrucomicrobiota (Akkermansiaceae) and Firmicutes (Lactobacillaceae) were still overwhelmingly dominant in gut microbiota after FMT in GF mice (Fig. 6b). Based on LEfSe analysis at the phylum and family levels (LDA score > 3.5), these were increased abundances of Bacteroidota and Muribaculaceae and a decreased abundances of Firmicutes and Lactobacillaceae in Tre-FMT-SAP group compared to BC-FMT-SAP group (Fig. 6c). These findings are consistent with the results of oral trehalose, further confirming that Muribaculaceae was the key factor in the role of trehalose.
Next, there was also a significant difference in gut microbial metabolism between BC-FMT-SAP and Tre-FMT-SAP groups (Fig. 6d, e). From ChemRICH class enrichment analysis of 134 differential metabolites (VIP ≥ 1), the abundances of organooxygen compounds & carboxylic acids and derivatives & glycerophospholipids were increased while the fatty acyls decreased in Tre-FMT-SAP group compared to BC-FMT-SAP group (Fig. 6f). For the 19 significantly different metabolites in Tre-FMT-SAP vs. BC-FMT-SAP group (VIP ≥ 1 and p < 0.05), the most significant pathway enrichment was that of galactose metabolism (Fig. 6g, h). These data confirm that galactose metabolism in the gut was a key factor in the role of trehalose.
To clarify whether there is a correlation between Muribaculaceae and Lactobacillaceae and significantly different metabolites, a network heatmap was performed. Muribaculaceae was positively correlated with these metabolites, while Lactobacillaceae was negatively correlated, and D-(+)-galactose (galactose metabolism) and 16-hydroxy hexadecanoic acid (biosynthesis of unsaturated fatty acids) were negatively correlated, which was consistent with the results for oral trehalose (Fig. 6i). Thus, we propose that galactose metabolism and Muribaculaceae in gut microbiota are key factors in trehalose-mediated “microbial metabolism-gut-pancreatic axis”.
Trehalose reduces macrophage infiltration and caspase-3-mediated apoptosis in the pancreas
The innate immune response, dominated by macrophages and neutrophils, is closely related to the severity of acute pancreatitis22. F4/80 and MPO expressions were detected in the pancreas. After ATBx treatment, immunohistochemistry (IHC) and immunofluorescence (IF) showed that pancreatic F4/80 and MPO expressions were decreased in ATBx-SAP group compared to SAP group (Fig. 7a). SAP-induced high expression of F4/80 was inhibited by oral trehalose or trehalose-remodeled gut microbiota, whereas MPO expression was not affected by oral trehalose or trehalose-remodeled gut microbiota (Fig. 7b-d). These data suggest that macrophage infiltration in the pancreas was decreased by trehalose, while neutrophil infiltration was not. Therefore, trehalose ameliorates SAP-induced inflammation by regulating macrophage.
The two programmed cell death pathways, autophagy and apoptosis, are considered to be associated with the progression of pancreatitis23. To determine whether the role of trehalose was related to cell apoptosis and autophagy, we examined the expression levels of Bcl-2, caspase-3, and Beclin1. Bcl-2, caspase-3, and Beclin1 expression were increased in the SAP group compared to the BC group (Fig. 8a-c). Notably, SAP-induced high expression of caspase-3 was significantly decreased by ATBx treatment, oral trehalose, and trehalose-remodelled gut microbiota, while Bcl-2 and Beclin1 expression were not affected (Fig. 8a-c). Based on the quantitative analysis of IHC and IF, SAP-induced high expression of caspase-3 was inhibited in the ATBx-SAP, Tre-SAP, and Tre-FMT-SAP group (Fig. 8d, e). Although the changes in Bcl-2 expression were not statistically significant (p > 0.05), there was a tendency toward increased Bcl-2 expression levels in the ATBx-SAP, Tre-SAP, and Tre-FMT-SAP group (Fig. 8d, e). Consistent with IHC and IF results, western blotting showed that SAP-induced high-cleaved caspase-3 expression was decreased by ATBx treatment, oral trehalose, and trehalose-remodelled gut microbiota while pro-caspase-3 expression was increased (Fig. 8f). These results suggest that trehalose ameliorates SAP-induced necrotic by regulated the caspase-3-mediated apoptosis.
Additionally, we found that Beclin1, caspase-3, and MPO were also expressed in pancreatic islets, with Beclin1 in α cells, caspase-3 in α/β cells, and MPO in β cells (Supplementary Fig. 6). Beclin1 and MPO levels in islets did not change during SAP progression, but caspase-3 expression in islets was increased in SAP group compared to BC group, which was inhibited by ATBx treatment. Oral trehalose or trehalose-remodeled gut microbiota did not affect SAP-induced caspase-3 expression increase in islet. These findings imply that gut microbiota are involved in islet injury during SAP progression.