Male diabetic mice (db/db), aged 10 wk and weighing 40-55 g, were purchased from GemPharmatech Co., Ltd. (Nanjing, Jiangsu, China) and housed at the Experimental Animal Center of Wenzhou Medical University (Zhejiang, China). All experimental protocols involving animals were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Committee on Animal Health and Care of Wenzhou Medical University. Before the experiment, all animals were provided with a normal diet and allowed to acclimatize for 1 wk under a 12:12 Light-dark cycle at room temperature (23 ± 1°C) and approximately 60% humidity. Before AP modeling, the animals were subjected to a 12-h fasting period and had ad libitum access to drinking water.

Mice with fasting blood glucose levels > 16.7 mmol/L were regarded as diabetic mice and were randomly divided into the following groups (n = 5 per group): Diabetic mouse group (db), ceruletide-induced AP model group (AP), FGF21 treatment group (FGF21), and FGF21 combined with Abx cocktail treatment group (FGF21 + Abx). AP model was established in mice of the AP, FGF21, and FGF21 + Abx groups, wherein each mouse received seven intraperitoneal injections of ceruletide (50 μg/kg), at hourly intervals[13,14]. The mice in the db group received intraperitoneal injections of the same volume of normal saline as a control. The successful establishment of the AP mouse model was confirmed based on the following criteria: (1) Increased activity of serum AMS released from the pancreas, wherein the enzyme was detected using an enzyme-linked immunosorbent assay kit after AP induction; (2) no increase in pancreatic AMS levels; and (3) pancreatic tissues not meeting the diagnostic criteria for pancreatitis according to the modified Schmidt scoring system.

The animals were euthanized 6 h after the final injection of ceruletide. Serum and pancreatic and intestinal tissues were collected from mice of each group to determine the serum AMS levels and the ratio of pancreas weight to body weight. The collected tissues were embedded in paraffin, sliced, and stained with hematoxylin and eosin (HE), followed by a microscopic examination to observe the morphological changes in pancreatic tissues. 16S rRNA sequencing was performed to observe changes in the gut microbiota.

Experimental mice in the FGF21 and FGF21 + Abx groups received an intraperitoneal injection of FGF21 (1 mg/kg) 1 h before the ceruletide injection. In the same manner, mice from the db and AP groups received intraperitoneal injections of normal saline as a control.

Mice in the FGF21 + Abx group were orally administered with an Abx cocktail of non-absorbable Abx (ampicillin, neomycin, metronidazole, and vancomycin). The Abx cocktail was prepared at concentrations of 1 g/L, 1 g/L, 1 g/L, and 0.5 mg/L for each of these Abx, respectively. The Abx cocktail solution was freshly prepared every 2 d and administered continuously for 3 wk.

After 3 wk of Abx cocktail treatment, fecal samples were collected from mice in the FGF21 + Abx group for fecal DNA extraction[13], followed by the detection of bacteria in the intestines using universal primers[14]. After depletion of the majority of bacteria in the mouse intestine, the mice in the FGF21 + Abx group received an intraperitoneal injection of FGF21, followed by an intraperitoneal injection of ceruletide 1 h later to establish the AP model in diabetic mice.

Total protein was extracted from the pancreatic and intestinal tissues of mice, and the protein concentration was determined using a bicinchoninic acid assay. The protein samples were separated through sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA). The membranes were then blocked with 10% skim milk–Tris-buffered saline with Tween 20 (TBST) solution at room temperature for 1.5 h, followed by an overnight incubation at 4 °C with the corresponding primary antibodies (IL-6 antibody diluted at 1:1000, IL-1β antibody diluted at 1:1000, TNF- antibody diluted at 1:2000, β-actin antibody diluted at 1:5000, and GAPDH antibody diluted at 1:5000, all purchased from Proteintech). The membranes were rinsed with TBST solution in triplicate and then incubated with the corresponding secondary antibodies conjugated with horseradish peroxidase at room temperature for 1 h. Chemiluminescent signals were detected using the Tanon-5200 chemiluminescence imaging system. The signal from each protein band was quantified using the ImageJ software.

Before histological analysis, mouse pancreatic and small intestinal tissues were fixed with 4% paraformaldehyde for more than 24 h. The fixed tissues were then embedded in paraffin, sliced at a thickness of 5 µm, and subjected to HE staining. The stained tissue sections were mounted with neutral resin and observed under a light microscope. The modified Schmidt scoring system was applied for the quantitative evaluation of pancreatic tissue damage.

The diversity of the gut microbiota in clinical or laboratory animal samples was analyzed using 16S rRNA sequencing and the next-generation microbiome bioinformatics platform QIIME 2[15]. The latest version of the QIIME 2 platform, together with the DADA2 software package, was used to denoise the sequence data using approximately 100% similarity, with an operational taxonomic unit (OTU) clustering at 97% similarity[16]. Redundancy was then removed to obtain feature data (representative sequences) for comparison with the 16S database (132 version) and NT-16. This comparison aimed to identify and annotate all 16S rRNA sequences detected in the samples, including taxonomic categories of kingdom, phylum, class, order, family, genus, and species.

Statistical analysis of data was performed using GraphPad Prism 6.0 software (GraphPad Software, San Diego, CA). Data were presented as mean ± SEM. Statistical significance was determined using Student's t-test (for comparisons between two experimental conditions) or analysis of variance (ANOVA) (for comparisons among three or more experimental conditions). Pearson analysis was used to determine linear correlations between variables. A P-value of less than 0.05 was considered to indicate statistical significance.

To investigate the impact of FGF21 on AP in diabetic mice, we established an AP model in diabetic mice by administering ceruletide injections. Following the ceruletide injection, the ratio of pancreas weight to body weight of diabetic mice was notably reduced by approximately 22.1% (Figure 1A, P < 0.01). Serum levels of AMS in diabetic mice were found to be twice as high as those in the db group (Figure 1B, P < 0.01). Concurrently, the concentration of the inflammatory cytokine IL-6 in diabetic mice increased to five times of that in the db group (Figure 1C and D, P < 0.01), while the TNF- level showed a significantly elevated to twice that of the db group (Figure 1C and E, P < 0.01). Following intraperitoneal injection of recombinant human FGF21 protein, the FGF21 group demonstrated an elevated ratio of pancreas weight to body weight, measuring at 19.3% (Figure 1A, P < 0.05). This result was accompanied by a reduction in the levels of serum AMS and inflammatory cytokines IL-6 and TNF- (Figure 1C). Specifically, the serum levels of AMS decreased by 40.1% (Figure 1B, P < 0.001), IL-6 levels decreased by 24.4% (Figure 1D, P < 0.01), and TNF- levels decreased to 65.1% of those in the AP group (Figure 1E, P < 0.05). Furthermore, diabetic mice with AP displayed pathological changes in pancreatic tissues, such as pancreatic edema, extensive intracellular vacuolation, and cellular necrosis (Figure 1F). Conversely, histological tissue sections of mice in the FGF21 group exhibited a significant reduction in tissue damage (Figure 1F).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Fibroblast growth factor 21 significantly attenuates pancreatic injury and inflammation in diabetic mice with acute pancreatitis. A: The ratio of pancreas weight to body weight of mice in diabetic (db), acute pancreatitis (AP) and fibroblast growth factor 21 (FGF21) groups; B: Serum levels of amylase of mice in db, AP and FGF21 groups; C-E: Representative immunoblots of inflammatory factors in mouse pancreatic tissue. Expression levels of interleukin-6 and tumor necrosis factor-alpha were quantified using densitometry, with GAPDH as a protein loading control; F: Pathological changes in pancreatic tissues of mice in db, AP and FGF21 groups, such as pancreatic edema, extensive intracellular vacuolation, and cellular necrosis (scale bar: 100 µm). Data are presented as mean ± SD, ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. AP: Acute pancreatitis; AMS: Amylase; db: Diabetic; IL: Interleukin; TNF-: Tumor necrosis factor-alpha; FGF21: Fibroblast growth factor 21.

These findings highlight that ceruletide injections induce pancreatic injury and inflammation in diabetic mice, while FGF21 treatment mitigates these symptoms in diabetic mice with AP.

To explore gut microbiota alterations in the context of AP, we assessed intestinal tissue damage in mice. The mice in the AP group showed exacerbated histologic evidence of inflammation in the small intestine, characterized by tissue edema, increased villus width, and villus damage (Figure 2A). After FGF21 treatment, noticeable reductions in the levels of inflammatory factors were observed in the small intestinal tissue (Figure 2B), with TNF- levels decreasing to 38.4% of that in the AP group (Figure 2C, P < 0.01), IL-6 levels decreasing by half (Figure 2D, P < 0.05), and IL-1β levels decreasing to 24.2% of that in the AP group (Figure 2E, P < 0.01), indicating a significant alleviation of intestinal tissue damage.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Fibroblast growth factor 21 treatment mitigates intestinal damage and inflammation in diabetic mice with acute pancreatitis. A: Histological changes in the small intestine of mice in diabetic, acute pancreatitis and fibroblast growth factor 21 groups, characterized by tissue edema, increased villus width, and villus damage (scale bar: 100 µm); B-E: Representative immunoblots of inflammatory factors in mouse small intestinal tissue. Expression levels of interleukin-6 and tumor necrosis factor-alpha were quantified using densitometry, with GAPDH as a protein loading control. Data are presented as mean ± SD, ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. AP: Acute pancreatitis; db: Diabetic; IL: Interleukin; TNF-: Tumor necrosis factor-alpha; FGF21: Fibroblast growth factor 21.

Next, 16S rRNA sequencing was performed to examine whether FGF21 altered the composition of the gut microbiota in diabetic mice while alleviating intestinal damage. Principal coordinate analysis (PCoA) results demonstrated distinct segregation of the microbial communities among the db, AP, and FGF21 groups, underscoring differences in the gut microbiota composition among the three groups (Figure 3A, P < 0.01). Alpha diversity of the gut microbiota, which serves as a comprehensive index of species abundance and evenness in community ecology, was assessed using four commonly used indices: observed OTUs, Chao1, Shannon, and Simpson indices. The observed OTUs and Chao1 indices reflect the species abundance in a sample, while the Shannon and Simpson indices reflect both the species abundance and evenness. When compared with the db group, all four indices showed significant increases in the AP group (P < 0.001). The FGF21 group exhibited notable decreases in OTUs and Chao1 indices when compared with the AP group (P < 0.05), with Shannon and Simpson indices showing a nonsignificant decrease (Figure 3B-E). These results indicate an increase in gut microbiota abundance in diabetic mice with AP, and FGF21 treatment could reverse this change. The gut microbiota plays a pivotal role in the occurrence of AP in diabetic mice.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Fibroblast growth factor 21 treatment improves the composition of gut microbiota. A: Principal coordinate analysis results demonstrated distinct segregation of the microbial communities among the diabetic (db), acute pancreatitis (AP), and fibroblast growth factor 21 (FGF21) groups. Different colors in the scatter plots represent samples from different groups; the higher the similarity between samples, the closer they are in the plots; B-E: The observed operational taxonomic units, Chao1, Shannon, and Simpson indices of the gut microbiota of db, AP and FGF21 group mice. All four indices increased in the AP group compared with the db group, and the FGF21 group exhibited decreases compared with the AP group; F: Bar graph of the structural distributions of fecal microbial communities at the phylum level; G: Relative abundance of the dominant phyla. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001, ^dP < 0.0001. AP: Acute pancreatitis; db: Diabetic; IL: Interleukin; TNF-: Tumor necrosis factor-alpha; FGF21: Fibroblast growth factor 21.

We further compared the alterations in bacterial communities among the three groups, with the stacked bar chart illustrating bacterial species distribution and changes in species composition and distribution within each group. The experimental results highlighted distinct variations in the highest relative abundance of the top 30 species at different taxonomic levels of the samples. At the phylum level, Firmicutes, Proteobacteria, Bacteroidetes, and Actinobacteria emerged as major taxa in abundance across all samples (Figure 3F). Notably, phylum Firmicutes exhibited a marked elevation in the AP group (Figure 3G, P < 0.01), with significant decreases in the FGF21 group (Figure 3G, P < 0.001). Meanwhile, phyla Proteobacteria and Bacteroidetes showed significant decreases in the AP group (Figure 3G, P < 0.05) and significant increases in the FGF21 group (Figure 3G, P < 0.05). Moreover, Firmicutes was found to be the most abundant phylum across all samples. This study found that the Bacteroides/Firmicutes ratio decreased in the AP group, with a rebound in the FGF21 group.

Upon disrupting the gut microbiota through an Abx cocktail, a notable decrease in bacterial abundance was observed in the mouse feces (Figure 4A). Following the commencement of Abx cocktail treatment, the body weight of mice decreased, reaching the lowest point on day 13, with an average body weight of 45 g. However, following adaptation to the Abx cocktail feeding, the body weight of mice gradually increased and reached 48 g (Figure 4B). In subsequent experiments, when compared with the AP group, the FGF21 + Abx group demonstrated a significant decrease in serum AMS levels (Figure 4C, P < 0.0001), with minimal damage observed in pancreatic and intestinal tissue sections (Figure 4D and E). Immunoblotting analysis revealed further reductions in the levels of the pro-inflammatory cytokines TNF- and IL-6 in the FGF21 + Abx group in the pancreatic tissue when compared with those in the FGF21 group (Figure 4F). Specifically, TNF- exhibited a significant decrease of 75.9% (Figure 4G, P < 0.01), and IL-6 showed a reduction to 25.8% of the FGF21 + Abx group (Figure 4H, P < 0.01). Similarly, noticeable reductions in the levels of inflammatory factors were observed in the small intestinal tissue (Figure 4I), with TNF- levels decreasing to 23.4% of that in the FGF21 group (Figure 4J, P < 0.001), IL-1β levels decreasing by half (Figure 4K, P < 0.05), and IL-6 levels decreasing to 45.6% (Figure 4L, P < 0.01). Notably, blood glucose levels significantly decreased from 17.50 ± 1.1 to 9.817 ± 0.69 mmol/L (Figure 4F, P < 0.001) in the FGF21 + Abx group, further decreasing from 15.14 ± 1.8 mmol/L in the FGF21 group (Figure 4M, P < 0.05). These experimental results demonstrate that the combination therapy of Abx cocktail with FGF21 exerts a more potent protective effect on AP in diabetic mice compared to FGF21 treatment alone. The Abx cocktail enhances the protective efficacy of FGF21 in diabetic mice with AP.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Combined therapy of Abx cocktail and fibroblast growth factor 21 significantly decreases the susceptibility to acute pancreatitis in diabetic mice. A: Verification of the intestinal microbiota removal after feeding antibiotics; B: Following Abx cocktail treatment initiation, the body weight change of mice; C: Serum amylase levels of mice in each group; D: Pathological changes in pancreatic tissues of mice in each group, such as pancreatic edema, extensive intracellular vacuolation, and cellular necrosis (scale bar: 100 µm); E: Histological changes in the small intestine of mice in each group, characterized by tissue edema, increased villus width, and villus damage (scale bar: 100 µm); F-H: Representative immunoblots of inflammatory factors in mouse pancreatic tissue. Expression levels of interleukin (IL)-6 and tumor necrosis factor (TNF)-alpha were quantified using densitometry, with GAPDH as a protein loading control; I-L: Representative immunoblots of inflammatory factors in mouse small intestinal tissue. Expression levels of IL-6, TNF- and IL-1 were quantified using densitometry, with GAPDH as a protein loading control. M: Changes in blood glucose in mice in different groups. Data are presented as mean ± SD, ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. AP: Acute pancreatitis; db: Diabetic; IL: Interleukin; TNF-: Tumor necrosis factor-alpha; FGF21: Fibroblast growth factor 21.

To further investigate the contribution of the microbiota to the protective effects of FGF21 in diabetic mice with AP, we compared the changes in bacterial communities among four groups of mice (Figure 5A and B). At the phylum level, Firmicutes, Proteobacteria, Bacteroidetes, and Actinobacteria were the major taxa of the bacterial communities across all four groups, accounting for 96.6% of the total abundance. In comparison with the db group, phylum Firmicutes showed a substantial increase to 66.2% in the AP group (Figure 5B, P < 0.01), while it decreased to 43.7% in the FGF21 group, reaching normal levels, and further decreased to 5.7% in the FGF21 + Abx group. Relative to the db group, phyla Proteobacteria and Bacteroidetes significantly decreased to 10.9% and 20.7%, respectively, in the AP group (Figure 5B, P < 0.05). After FGF21 treatment, the abundance of Proteobacteria and Bacteroidetes significantly increased to 18.9% and 35.5%, respectively, in the FGF21 group (Figure 5B, P < 0.05), reaching normal levels. In contrast, in the FGF21 + Abx group, the phylum Proteobacteria significantly increased to 87.3% (Figure 5B, P < 0.001), while the phylum Bacteroidetes significantly decreased to 1.3% (Figure 5B, P < 0.001). Additionally, significant differences in gut microbiota composition were observed at the genus level among different groups (Figure 5C and D). The predominant taxa in the gut microbiota of the four groups of mice were Lactobacillus, Mucispirillum-Klebsiella, and Escherichia coli-Shigella, accounting for 44.3% in the FGF21 + Abx group. Moreover, the AP group exhibited the highest abundance of Lactobacillus, accounting for 53%. In the FGF21 + Abx group, Escherichia coli-Shigella accounted for 42%, while its abundance remained below 1% in the other three groups. Notably, prior research has highlighted an increase in the abundance of the phylum Firmicutes (gram-positive bacteria) and a decrease in the abundance of the phylum Bacteroidetes in obese mice[13]. After FGF21 treatment, the proportion of Firmicutes was significantly reduced, while that of Bacteroidetes was significantly increased in diabetic mice with AP, underscoring the effectiveness of FGF21 in alleviating diabetes conditions in diabetic mice with AP.

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Abx cocktail combined with fibroblast growth factor 21 treatment alters microbiota in diabetic mice with acute pancreatitis. A: Bar graph of the structural distributions of fecal microbial communities at the phylum level; B: Relative abundance of the dominant phyla; C: Bar graph of the structural distributions of fecal microbial communities at the genus level; D: Relative abundance of the dominant genera; E: Linear discriminant analysis (LDA) scores for the differentially abundant bacterial taxa between each group (LDA > 4.5); F: The phylogenetic tree shows the origins of the microbiota at different taxonomic levels. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001. AP: Acute pancreatitis; db: Diabetic; FGF21: Fibroblast growth factor 21.

Through linear discriminant analysis (LDA) effect size (LEfSe), the biomarkers were compared among the four groups of samples. In this study, a threshold of > 4.5 LDA score and P < 0.05 were set for the LEfSe analysis. The variance in LDA scores across the four groups of microbiota samples revealed that the phylum Proteobacteria, order Enterobacteriales, and family Enterobacteriaceae were the most abundant species in the FGF21 + Abx group. Escherichia coli-Shigella was found to be the most abundant species in the FGF21 group. The phylum Firmicutes, class Clostridia, and order Clostridiales were most abundant in the db group. These bacteria may serve as potential targets for the treatment of diabetes complicated with AP. The order Lactobacillales, family Lactobacillaceae, and genus Lactobacillus were the most abundant in the AP group (Figure 5E). These bacteria belong to the class Bacilli. The phylogenetic tree in Figure 5F shows the origins of the microbiota at different taxonomic levels.

PICRUSt was used to predict the potential functions of microbial genes[17]. In this study, we adopted the PICRUSt2 prediction method to obtain gene function annotations from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Subsequently, statistical analysis of the metagenomic profiles (STAMP) was utilized for differential expression analysis[18] to identify significantly different gene functions among groups. Of note, the KEGG pathway database integrates current knowledge of molecular interaction networks, including biochemical processes such as metabolism, membrane translocation, signal transduction, cell cycle, and conserved subpathways in the same cell lineage.

The obtained findings suggested that the differential pathways (P < 0.05) between the FGF21 + Abx group and the AP group included toluene degradation I (aerobic), toluene degradation III (aerobic), sulfate reduction I, cob(II) acetate a, c-diamine biosynthesis I, and the superpathway of n-acetylceramide degradation (Figure 6A). In addition, the differential pathways (P < 0.05) between the db and AP groups included histidine, purine, and pyrimidine biosynthesis, methoxy-13 biosynthesis, methoxy-12 biosynthesis, methoxy-11 biosynthesis, and methoxy-8 biosynthesis (Figure 6B). Furthermore, differential pathways (P < 0.05) between the db and FGF21 + Abx groups included the superpathway of sulfate assimilation and cysteine biosynthesis, superpathway of L-alanine biosynthesis, glyoxylate cycle, tricarboxylic acid cycle I prokaryote, and fatty acid β-oxidation I (Figure 6C). These distinct differentially expressed pathways may provide critical insights into the effects of AP treatment.

Open in New Tab   Full Size Figure   Download Figure

Figure 6 The significantly different gene functions among groups. A: The differential pathways between the fibroblast growth factor 21 (FGF21) + Abx group and the acute pancreatitis (AP) group; B: The differential pathways between the diabetic (db) and AP groups; C; The differential pathways between the db and FGF21 + Abx groups. AP: Acute pancreatitis; db: Diabetic; FGF21: Fibroblast growth factor 21.

Fibroblast growth factor 21 (FGF21) plays a pivotal role in regulating glucose and lipid metabolism. Acute pancreatitis (AP) is a common inflammatory disease with clinical manifestations. Diabetes exacerbates intestinal permeability and intestinal inflammation, thus leading to the progression to AP. Our previous study indicated that FGF21 significantly attenuated susceptibility to AP in mice.

Yet, whether FGF21 similarly protects AP in diabetic mice remains unexplored.

Herein, we were intrigued to investigate the potential protective role of FGF21 against AP in diabetic mice.

In the present study, a mouse model of AP was established in db/db diabetic mice through ceruletide injections. By comparing the differences in AP indicators between diabetic mouse group (db), ceruletide-induced AP model group (AP), FGF21 treatment group (FGF21), and FGF21 combined with an antibiotic (Abx) cocktail treatment group (FGF21 + Abx), we investigated the protective effect of recombinant FGF21 protein and investigated whether FGF21 plays its role in the treatment of diabetic mice with AP by modulating the gut microbiota.

FGF21 notably diminished the levels of serum amylase, inflammatory factors and the histological evidence of inflammation in the pancreas and the small intestine in diabetic mice with AP. FGF21 also significantly altered the composition of the gut microbiota, reestablishing the Bacteroidetes/Firmicutes ratio. Upon treatment with an Abx cocktail to deplete the gut microbiota, the FGF21 + Abx group showed superior protective effect. The gut microbiota composition across different groups was further characterized, and a differential expression analysis of gene functions was undertaken using the PICRUSt2 prediction method. These findings suggested that FGF21 could potentially confer therapeutic effects on diabetic mice with AP by modulating the sulfate reduction I pathway and the superpathway of n-acetylceramide degradation in the gut microbiota.

This study reveals the potential of FGF21 in improving pancreatic and intestinal damage recovery, reducing blood glucose levels, and reshaping gut microbiota composition in diabetic mice with AP. Notably, the protective effects of FGF21 are augmented when combined with the Abx cocktail. These findings provide new insights into the prevention and treatment of diabetes complicated by AP.

Further investigation is required to elucidate the specific mechanisms by which the gut microbiota affects the protective effects of FGF21 against AP in diabetic mice.